NETWORK DEVICE

RELATED APPLICATION

[0001] The present application claims the benefit of priority to U.S. Patent Application No. 17/103,602, filed November 24, 2020, entitled, “PAGE-BASED REMOTE MEMORY ACCESS USING SYSTEM MEMORY INTERFACE NETWORK DEVICE”, which in turn claims the benefit of priority to U.S. Provisional Application No. 63/115,511, filed November 18, 2020, entitled “SHARED MEMORY”, the entire contents of which are incorporated herein by reference.

DESCRIPTION

[0002] Scale-out and distributed architectures increase computing resources or available memory or storage by adding processors, memory, and storage for access using a fabric or network. Disaggregated memory architectures can use pools of memory, located remote from the compute nodes in the system. A memory pool can be shared across a rack or set of racks in a data center. As the memory pools are remote from a compute node, there are additional latencies inherent in their accesses due to latency of packet formation and processing and media traversal across one or more network elements.

BRIEF DESCRIPTION OF THE DRAWINGS

[0003] FIG. 1 is a simplified diagram of at least one embodiment of a data center for executing workloads with disaggregated resources.

[0004] FIG. 2 is a simplified diagram of at least one embodiment of a pod that may be included in a data center.

[0005] FIG. 3 is a perspective view of at least one embodiment of a rack that may be included in a pod.

[0006] FIG. 4 is a side elevation view of a rack.

[0007] FIG. 5 is a perspective view of a rack having a sled mounted therein.

[0008] FIG. 6 is a simplified block diagram of at least one embodiment of a top side of a sled.

[0009] FIG. 7 is a simplified block diagram of at least one embodiment of a bottom side of a sled.

[0010] FIG. 8 is a simplified block diagram of at least one embodiment of a compute sled.

[0011] FIG. 9 is atop perspective view of at least one embodiment of a compute sled.

[0012] FIG. 10 is a simplified block diagram of at least one embodiment of an accelerator sled usable in a data center.

[0013] FIG. 11 is atop perspective view of at least one embodiment of an accelerator sled.

[0014] FIG. 12 is a simplified block diagram of at least one embodiment of a storage sled usable in a data center.

[0015] FIG. 13 is atop perspective view of at least one embodiment of a storage sled. [0016] FIG. 14 is a simplified block diagram of at least one embodiment of a memory sled usable in a data center.

[0017] FIG. 15 depicts a system for executing one or more workloads.

[0018] FIG. 16 depicts an example of memory pools accessible by one or more hosts.

[0019] FIG. 17 depicts an example system.

[0020] FIG. 18 depicts an example of message transmission ordering.

[0021] FIGs. 19A and 19B depict examples of commanding a DMA engine.

[0022] FIG. 20 depicts an example process.

[0023] FIG. 21 depicts an example system.

[0024] FIG. 22 depicts an example sequence.

[0025] FIG. 23 depicts an example process.

[0026] FIG. 24 depicts an example network interface.

[0027] FIG. 25 depicts an example system.

DETAILED DESCRIPTION

[0028] FIG. 1 depicts a data center in which disaggregated resources may cooperatively execute one or more workloads (e.g., applications on behalf of customers) includes multiple pods 110, 120, 130, 140, a pod being or including one or more rows of racks. Of course, although data center 100 is shown with multiple pods, in some embodiments, the data center 100 may be embodied as a single pod. As described in more detail herein, each rack houses multiple sleds, each of which may be primarily equipped with a particular type of resource (e.g., memory devices, data storage devices, accelerator devices, general purpose processors), e.g., resources that can be logically coupled to form a composed node, which can act as, for example, a server. In the illustrative embodiment, the sleds in each pod 110, 120, 130, 140 are connected to multiple pod switches (e.g., switches that route data communications to and from sleds within the pod). The pod switches, in turn, connect with spine switches 150 that switch communications among pods (e.g., the pods 110, 120, 130, 140) in the data center 100. In some embodiments, the sleds may be connected with a fabric using Intel® Omni-Path technology. In other embodiments, the sleds may be connected with other fabrics, such as InfiniBand or Ethernet. As described in more detail herein, resources within sleds in the data center 100 may be allocated to a group (referred to herein as a “managed node”) containing resources from one or more sleds to be collectively utilized in the execution of a workload. The workload can execute as if the resources belonging to the managed node were located on the same sled. The resources in a managed node may belong to sleds belonging to different racks, and even to different pods 110, 120, 130, 140. As such, some resources of a single sled may be allocated to one managed node while other resources of the same sled are allocated to a different managed node (e.g., one processor assigned to one managed node and another processor of the same sled assigned to a different managed node).

[0029] A data center comprising disaggregated resources, such as data center 100, can be used in a wide variety of contexts, such as enterprise, government, cloud service provider, and communications service provider (e.g., Telco’s), as well in a wide variety of sizes, from cloud service provider mega-data centers that consume over 100,000 sq. ft. to single- or multi-rack installations for use in base stations.

[0030] The disaggregation of resources to sleds comprised predominantly of a single type of resource (e.g., compute sleds comprising primarily compute resources, memory sleds containing primarily memory resources), and the selective allocation and deallocation of the disaggregated resources to form a managed node assigned to execute a workload improves the operation and resource usage of the data center 100 relative to typical data centers comprised of hyperconverged servers containing compute, memory, storage and perhaps additional resources in a single chassis. For example, because sleds predominantly contain resources of a particular type, resources of a given type can be upgraded independently of other resources. Additionally, because different resources types (processors, storage, accelerators, etc.) typically have different refresh rates, greater resource utilization and reduced total cost of ownership may be achieved. For example, a data center operator can upgrade the processors throughout their facility by only swapping out the compute sleds. In such a case, accelerator and storage resources may not be contemporaneously upgraded and, rather, may be allowed to continue operating until those resources are scheduled for their own refresh. Resource utilization may also increase. For example, if managed nodes are composed based on requirements of the workloads that will be running on them, resources within a node are more likely to be fully utilized. Such utilization may allow for more managed nodes to run in a data center with a given set of resources, or for a data center expected to run a given set of workloads, to be built using fewer resources.

[0031] FIG. 2 depicts a pod. A pod can include a set of rows 200, 210, 220, 230 of racks 240. Each rack 240 may house multiple sleds (e.g., sixteen sleds) and provide power and data connections to the housed sleds, as described in more detail herein. In the illustrative embodiment, the racks in each row 200, 210, 220, 230 are connected to multiple pod switches 250, 260. The pod switch 250 includes a set of ports 252 to which the sleds of the racks of the pod 110 are connected and another set of ports 254 that connect the pod 110 to the spine switches 150 to provide connectivity to other pods in the data center 100. Similarly, the pod switch 260 includes a set of ports 262 to which the sleds of the racks of the pod 110 are connected and a set of ports 264 that connect the pod 110 to the spine switches 150. As such, the use of the pair of switches 250, 260 provides an amount of redundancy to the pod 110. For example, if either of the switches 250, 260 fails, the sleds in the pod 110 may still maintain data communication with the remainder of the data center 100 (e.g., sleds of other pods) through the other switch 250, 260. Furthermore, in the illustrative embodiment, the switches 150, 250, 260 may be embodied as dual-mode optical switches, capable of routing both Ethernet protocol communications carrying Internet Protocol (IP) packets and communications according to a second, high-performance link-layer protocol (e.g., PCI Express) via optical signaling media of an optical fabric.

[0032] It should be appreciated that each of the other pods 120, 130, 140 (as well as any additional pods of the data center 100) may be similarly structured as, and have components similar to, the pod 110 shown in and described in regard to FIG. 2 (e.g., each pod may have rows of racks housing multiple sleds as described above). Additionally, while two pod switches 250, 260 are shown, it should be understood that in other embodiments, each pod 110, 120, 130, 140 may be connected to a different number of pod switches, providing even more failover capacity. Of course, in other embodiments, pods may be arranged differently than the rows-of-racks configuration shown in FIGS. 1-2. For example, a pod may be embodied as multiple sets of racks in which each set of racks is arranged radially, e.g., the racks are equidistant from a center switch.

[0033] Referring now to FIGs. 3-5, each illustrative rack 240 of the data center 100 includes two elongated support posts 302, 304, which are arranged vertically. For example, the elongated support posts 302, 304 may extend upwardly from a floor of the data center 100 when deployed. The rack 240 also includes one or more horizontal pairs 310 of elongated support arms 312 (identified in FIG. 3 via a dashed ellipse) configured to support a sled of the data center 100 as discussed below. One elongated support arm 312 of the pair of elongated support arms 312 extends outwardly from the elongated support post 302 and the other elongated support arm 312 extends outwardly from the elongated support post 304.

[0034] In the illustrative embodiments, each sled of the data center 100 is embodied as a chassis-less sled. That is, each sled has a chassis-less circuit board substrate on which physical resources (e.g., processors, memory, accelerators, storage, etc.) are mounted as discussed in more detail below. As such, the rack 240 is configured to receive the chassis-less sleds. For example, each pair 310 of elongated support arms 312 defines a sled slot 320 of the rack 240, which is configured to receive a corresponding chassisless sled. To do so, each illustrative elongated support arm 312 includes a circuit board guide 330 configured to receive the chassis-less circuit board substrate of the sled. Each circuit board guide 330 is secured to, or otherwise mounted to, a top side 332 of the corresponding elongated support arm 312. For example, in the illustrative embodiment, each circuit board guide 330 is mounted at a distal end of the corresponding elongated support arm 312 relative to the corresponding elongated support post 302, 304. For clarity of the Figures, not every circuit board guide 330 may be referenced in each Figure.

[0035] Each circuit board guide 330 includes an inner wall that defines a circuit board slot 380 configured to receive the chassis-less circuit board substrate of a sled 400 when the sled 400 is received in the corresponding sled slot 320 of the rack 240. To do so, as shown in FIG. 4, a user (or robot) aligns the chassis-less circuit board substrate of an illustrative chassis-less sled 400 to a sled slot 320. The user, or robot, may then slide the chassis-less circuit board substrate forward into the sled slot 320 such that each side edge 414 of the chassis-less circuit board substrate is received in a corresponding circuit board slot 380 of the circuit board guides 330 of the pair 310 of elongated support arms 312 that define the corresponding sled slot 320 as shown in FIG. 4. By having robotically accessible and robotically manipulatable sleds comprising disaggregated resources, each type of resource can be upgraded independently of each other and at their own optimized refresh rate. Furthermore, the sleds are configured to blindly mate with power and data communication cables in each rack 240, enhancing their ability to be quickly removed, upgraded, reinstalled, and/or replaced. As such, in some embodiments, the data center 100 may operate (e.g., execute workloads, undergo maintenance and/or upgrades, etc.) without human involvement on the data center floor. In other embodiments, a human may facilitate one or more maintenance or upgrade operations in the data center 100. [0036] It should be appreciated that each circuit board guide 330 is dual sided. That is, each circuit board guide 330 includes an inner wall that defines a circuit board slot 380 on each side of the circuit board guide 330. In this way, each circuit board guide 330 can support a chassis-less circuit board substrate on either side. As such, a single additional elongated support post may be added to the rack 240 to turn the rack 240 into a two-rack solution that can hold twice as many sled slots 320 as shown in Fig. 3. The illustrative rack 240 includes seven pairs 310 of elongated support arms 312 that define a corresponding seven sled slots 320, each configured to receive and support a corresponding sled 400 as discussed above. Of course, in other embodiments, the rack 240 may include additional or fewer pairs 310 of elongated support arms 312 (e.g., additional or fewer sled slots 320). It should be appreciated that because the sled 400 is chassis-less, the sled 400 may have an overall height that is different than typical servers. As such, in some embodiments, the height of each sled slot 320 may be shorter than the height of a typical server (e.g., shorter than a single rank unit, “1U”). That is, the vertical distance between each pair 310 of elongated support arms 312 may be less than a standard rack unit “1U.” Additionally, due to the relative decrease in height of the sled slots 320, the overall height of the rack 240 in some embodiments may be shorter than the height of traditional rack enclosures. For example, in some embodiments, each of the elongated support posts 302, 304 may have a length of six feet or less. Again, in other embodiments, the rack 240 may have different dimensions. For example, in some embodiments, the vertical distance between each pair 310 of elongated support arms 312 may be greater than a standard rack until “1U” In such embodiments, the increased vertical distance between the sleds allows for larger heat sinks to be attached to the physical resources and for larger fans to be used (e.g., in the fan array 370 described below) for cooling each sled, which in turn can allow the physical resources to operate at increased power levels. Further, it should be appreciated that the rack 240 does not include any walls, enclosures, or the like. Rather, the rack 240 is an enclosure-less rack that is opened to the local environment. Of course, in some cases, an end plate may be attached to one of the elongated support posts 302, 304 in those situations in which the rack 240 forms an end-of-row rack in the data center 100.

[0037] In some embodiments, various interconnects may be routed upwardly or downwardly through the elongated support posts 302, 304. To facilitate such routing, each elongated support post 302, 304 includes an inner wall that defines an inner chamber in which interconnects may be located. The interconnects routed through the elongated support posts 302, 304 may be embodied as any type of interconnects including, but not limited to, data or communication interconnects to provide communication connections to each sled slot 320, power interconnects to provide power to each sled slot 320, and/or other types of interconnects.

[0038] The rack 240, in the illustrative embodiment, includes a support platform on which a corresponding optical data connector (not shown) is mounted. Each optical data connector is associated with a corresponding sled slot 320 and is configured to mate with an optical data connector of a corresponding sled 400 when the sled 400 is received in the corresponding sled slot 320. In some embodiments, optical connections between components (e.g., sleds, racks, and switches) in the data center 100 are made with a blind mate optical connection. For example, a door on each cable may prevent dust from contaminating the fiber inside the cable . In the process of connecting to a blind mate optical connector mechanism, the door is pushed open when the end of the cable approaches or enters the connector mechanism. Subsequently, the optical fiber inside the cable may enter a gel within the connector mechanism and the optical fiber of one cable comes into contact with the optical fiber of another cable within the gel inside the connector mechanism.

[0039] The illustrative rack 240 also includes a fan array 370 coupled to the cross-support arms of the rack 240. The fan array 370 includes one or more rows of cooling fans 372, which are aligned in a horizontal line between the elongated support posts 302, 304. In the illustrative embodiment, the fan array 370 includes a row of cooling fans 372 for each sled slot 320 of the rack 240. As discussed above, each sled 400 does not include any on-board cooling system in the illustrative embodiment and, as such, the fan array 370 provides cooling for each sled 400 received in the rack 240. Each rack 240, in the illustrative embodiment, also includes a power supply associated with each sled slot 320. Each power supply is secured to one of the elongated support arms 312 of the pair 310 of elongated support arms 312 that define the corresponding sled slot 320. For example, the rack 240 may include a power supply coupled or secured to each elongated support arm 312 extending from the elongated support post 302. Each power supply includes a power connector configured to mate with a power connector of the sled 400 when the sled 400 is received in the corresponding sled slot 320. In the illustrative embodiment, the sled 400 does not include any on-board power supply and, as such, the power supplies provided in the rack 240 supply power to corresponding sleds 400 when mounted to the rack 240. Each power supply is configured to satisfy the power requirements for its associated sled, which can vary from sled to sled. Additionally, the power supplies provided in the rack 240 can operate independent of each other. That is, within a single rack, a first power supply providing power to a compute sled can provide power levels that are different than power levels supplied by a second power supply providing power to an accelerator sled. The power supplies may be controllable at the sled level or rack level, and may be controlled locally by components on the associated sled or remotely, such as by another sled or an orchestrator.

[0040] Referring now to FIG. 6, the sled 400, in the illustrative embodiment, is configured to be mounted in a corresponding rack 240 of the data center 100 as discussed above. In some embodiments, each sled 400 may be optimized or otherwise configured for performing particular tasks, such as compute tasks, acceleration tasks, data storage tasks, etc. For example, the sled 400 may be embodied as a compute sled 800 as discussed below in regard to FIGS. 8-9, an accelerator sled 1000 as discussed below in regard to FIGS. 10-11, a storage sled 1200 as discussed below in regard to FIGS. 12-13, or as a sled optimized or otherwise configured to perform other specialized tasks, such as a memory sled 1400, discussed below in regard to FIG. 14.

[0041] As discussed above, the illustrative sled 400 includes a chassis-less circuit board substrate 602, which supports various physical resources (e.g., electrical components) mounted thereon. It should be appreciated that the circuit board substrate 602 is “chassis-less” in that the sled 400 does not include a housing or enclosure. Rather, the chassis-less circuit board substrate 602 is open to the local environment. The chassis-less circuit board substrate 602 may be formed from any material capable of supporting the various electrical components mounted thereon. For example, in an illustrative embodiment, the chassisless circuit board substrate 602 is formed from an FR-4 glass-reinforced epoxy laminate material. Of course, other materials may be used to form the chassis-less circuit board substrate 602 in other embodiments.

[0042] As discussed in more detail below, the chassis-less circuit board substrate 602 includes multiple features that improve the thermal cooling characteristics of the various electrical components mounted on the chassis-less circuit board substrate 602. As discussed, the chassis-less circuit board substrate 602 does not include a housing or enclosure, which may improve the airflow over the electrical components of the sled 400 by reducing those structures that may inhibit air flow. For example, because the chassis-less circuit board substrate 602 is not positioned in an individual housing or enclosure, there is no vertically-arranged backplane (e.g., a backplate of the chassis) attached to the chassis-less circuit board substrate 602, which could inhibit air flow across the electrical components. Additionally, the chassis-less circuit board substrate 602 has a geometric shape configured to reduce the length of the airflow path across the electrical components mounted to the chassis-less circuit board substrate 602. For example, the illustrative chassisless circuit board substrate 602 has a width 604 that is greater than a depth 606 of the chassis-less circuit board substrate 602. In one particular embodiment, for example, the chassis-less circuit board substrate 602 has a width of about 21 inches and a depth of about 9 inches, compared to a typical server that has a width of about 17 inches and a depth of about 39 inches. As such, an airflow path 608 that extends from a front edge 610 of the chassis-less circuit board substrate 602 toward a rear edge 612 has a shorter distance relative to typical servers, which may improve the thermal cooling characteristics of the sled 400. Furthermore, although not illustrated in FIG. 6, the various physical resources mounted to the chassis-less circuit board substrate 602 are mounted in corresponding locations such that no two substantively heatproducing electrical components shadow each other as discussed in more detail below. That is, no two electrical components, which produce appreciable heat during operation (e.g., greater than a nominal heat sufficient enough to adversely impact the cooling of another electrical component), are mounted to the chassis-less circuit board substrate 602 linearly in-line with each other along the direction of the airflow path 608 (e.g., along a direction extending from the front edge 610 toward the rear edge 612 of the chassisless circuit board substrate 602).

[0043] As discussed above, the illustrative sled 400 includes one or more physical resources 620 mounted to a top side 650 of the chassis-less circuit board substrate 602. Although two physical resources 620 are shown in FIG. 6, it should be appreciated that the sled 400 may include one, two, or more physical resources 620 in other embodiments. The physical resources 620 may be embodied as any type of processor, controller, or other compute circuit capable of performing various tasks such as compute functions and/or controlling the functions of the sled 400 depending on, for example, the type or intended functionality of the sled 400. For example, as discussed in more detail below, the physical resources 620 may be embodied as high-performance processors in embodiments in which the sled 400 is embodied as a compute sled, as accelerator co-processors or circuits in embodiments in which the sled 400 is embodied as an accelerator sled, storage controllers in embodiments in which the sled 400 is embodied as a storage sled, or a set of memory devices in embodiments in which the sled 400 is embodied as a memory sled.

[0044] The sled 400 also includes one or more additional physical resources 630 mounted to the top side 650 of the chassis-less circuit board substrate 602. In the illustrative embodiment, the additional physical resources include a network interface controller (NIC) as discussed in more detail below. Of course, depending on the type and functionality of the sled 400, the physical resources 630 may include additional or other electrical components, circuits, and/or devices in other embodiments.

[0045] The physical resources 620 are communicatively coupled to the physical resources 630 via an input/output (I/O) subsystem 622. The I/O subsystem 622 may be embodied as circuitry and/or components to facilitate input/output operations with the physical resources 620, the physical resources 630, and/or other components of the sled 400. For example, the I/O subsystem 622 may be embodied as, or otherwise include, memory controller hubs, input/output control hubs, integrated sensor hubs, firmware devices, communication links (e.g., point-to-point links, bus links, wires, cables, waveguides, light guides, printed circuit board traces, etc.), and/or other components and subsystems to facilitate the input/output operations. In the illustrative embodiment, the I/O subsystem 622 is embodied as, or otherwise includes, a double data rate 4 (DDR4) data bus or a DDR5 data bus.

[0046] In some embodiments, the sled 400 may also include a resource-to-resource interconnect 624. The resource-to-resource interconnect 624 may be embodied as any type of communication interconnect capable of facilitating resource-to-resource communications. In the illustrative embodiment, the resource- to-resource interconnect 624 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the resource-to-resource interconnect 624 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), PCI express (PCIe), or other high-speed point-to-point interconnect dedicated to resource-to-resource communications.

[0047] The sled 400 also includes a power connector 640 configured to mate with a corresponding power connector of the rack 240 when the sled 400 is mounted in the corresponding rack 240. The sled 400 receives power from a power supply of the rack 240 via the power connector 640 to supply power to the various electrical components of the sled 400. That is, the sled 400 does not include any local power supply (e.g., an on-board power supply) to provide power to the electrical components of the sled 400. The exclusion of a local or on-board power supply facilitates the reduction in the overall footprint of the chassisless circuit board substrate 602, which may increase the thermal cooling characteristics of the various electrical components mounted on the chassis-less circuit board substrate 602 as discussed above. In some embodiments, voltage regulators are placed on a bottom side 750 (see FIG. 7) of the chassis-less circuit board substrate 602 directly opposite of the processors 820 (see FIG. 8), and power is routed from the voltage regulators to the processors 820 by vias extending through the circuit board substrate 602. Such a configuration provides an increased thermal budget, additional current and/or voltage, and better voltage control relative to typical printed circuit boards in which processor power is delivered from a voltage regulator, in part, by printed circuit traces.

[0048] In some embodiments, the sled 400 may also include mounting features 642 configured to mate with a mounting arm, or other structure, of a robot to facilitate the placement of the sled 600 in a rack 240 by the robot. The mounting features 642 may be embodied as any type of physical structures that allow the robot to grasp the sled 400 without damaging the chassis-less circuit board substrate 602 or the electrical components mounted thereto. For example, in some embodiments, the mounting features 642 may be embodied as non-conductive pads attached to the chassis-less circuit board substrate 602. In other embodiments, the mounting features may be embodied as brackets, braces, or other similar structures attached to the chassis-less circuit board substrate 602. The particular number, shape, size, and/or makeup of the mounting feature 642 may depend on the design of the robot configured to manage the sled 400. [0049] Referring now to FIG. 7, in addition to the physical resources 630 mounted on the top side 650 of the chassis-less circuit board substrate 602, the sled 400 also includes one or more memory devices 720 mounted to a bottom side 750 of the chassis-less circuit board substrate 602. That is, the chassis-less circuit board substrate 602 is embodied as a double-sided circuit board. The physical resources 620 are communicatively coupled to the memory devices 720 via the I/O subsystem 622. For example, the physical resources 620 and the memory devices 720 may be communicatively coupled by one or more vias extending through the chassis-less circuit board substrate 602. Each physical resource 620 may be communicatively coupled to a different set of one or more memory devices 720 in some embodiments. Alternatively, in other embodiments, each physical resource 620 may be communicatively coupled to each memory device 720. [0050] The memory devices 720 may be embodied as any type of memory device capable of storing data for the physical resources 620 during operation of the sled 400, such as any type of volatile (e.g., dynamic random access memory (DRAM), etc.) or non-volatile memory. Volatile memory may be a storage medium that requires power to maintain the state of data stored by the medium. Non-limiting examples of volatile memory may include various types of random access memory (RAM), such as dynamic random access memory (DRAM) or static random access memory (SRAM). One particular type of DRAM that may be used in a memory module is synchronous dynamic random access memory (SDRAM). In particular embodiments, DRAM of a memory component may comply with a standard promulgated by JEDEC, such as JESD79F for DDR SDRAM, JESD79-2F for DDR2 SDRAM, JESD79- 3F for DDR3 SDRAM, JESD79-4A for DDR4 SDRAM, JESD209 for Low Power DDR (LPDDR), JESD209-2 for LPDDR2, JESD209-3 for LPDDR3, and JESD209-4 for LPDDR4. Such standards (and similar standards) may be referred to as DDR-based standards and communication interfaces of the storage devices that implement such standards may be referred to as DDR-based interfaces.

[0051] In one embodiment, the memory device is a block addressable memory device, such as those based on NAND or NOR technologies. A block can be any size such as but not limited to 2KB, 4KB, 8KB, and so forth. A memory device may also include next-generation nonvolatile devices, such as Intel Optane® memory or other byte addressable write-in-place nonvolatile memory devices. In one embodiment, the memory device may be or may include memory devices that use chalcogenide glass, multithreshold level NAND flash memory, NOR flash memory, single or multi-level Phase Change Memory (PCM), a resistive memory, nanowire memory, ferroelectric transistor random access memory (FeTRAM), anti-ferroelectric memory, magnetoresistive random access memory (MRAM) memory that incorporates memristor technology, resistive memory including the metal oxide base, the oxygen vacancy base and the conductive bridge Random Access Memory (CB-RAM), or spin transfer torque (STT)-MRAM, a spintronic magnetic junction memory based device, a magnetic tunneling junction (MTJ) based device, a DW (Domain Wall) and SOT (Spin Orbit Transfer) based device, a thyristor based memory device, or a combination of any of the above, or other memory. The memory device may refer to the die itself and/or to a packaged memory product. In some embodiments, the memory device may comprise a transistor-less stackable cross point architecture in which memory cells sit at the intersection of word lines and bit lines and are individually addressable and in which bit storage is based on a change in bulk resistance.

[0052] Referring now to FIG. 8, in some embodiments, the sled 400 may be embodied as a compute sled 800. The compute sled 800 is optimized, or otherwise configured, to perform compute tasks. Of course, as discussed above, the compute sled 800 may rely on other sleds, such as acceleration sleds and/or storage sleds, to perform such compute tasks. The compute sled 800 includes various physical resources (e.g., electrical components) similar to the physical resources of the sled 400, which have been identified in FIG. 8 using the same reference numbers. The description of such components provided above in regard to FIGS. 6 and 7 applies to the corresponding components of the compute sled 800 and is not repeated herein for clarity of the description of the compute sled 800.

[0053] In the illustrative compute sled 800, the physical resources 620 are embodied as processors 820. Although only two processors 820 are shown in FIG. 8, it should be appreciated that the compute sled 800 may include additional processors 820 in other embodiments. Illustratively, the processors 820 are embodied as high-performance processors 820 and may be configured to operate at a relatively high power rating. Although the processors 820 generate additional heat operating at power ratings greater than typical processors (which operate at around 155-230 W), the enhanced thermal cooling characteristics of the chassis-less circuit board substrate 602 discussed above facilitate the higher power operation. For example, in the illustrative embodiment, the processors 820 are configured to operate at a power rating of at least 250 W. In some embodiments, the processors 820 may be configured to operate at a power rating of at least 350 W.

[0054] In some embodiments, the compute sled 800 may also include a processor-to-processor interconnect 842. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the processor-to-processor interconnect 842 may be embodied as any type of communication interconnect capable of facilitating processor-to-processor interconnect 842 communications. In the illustrative embodiment, the processor-to-processor interconnect 842 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the processor-to-processor interconnect 842 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other highspeed point-to-point interconnect dedicated to processor-to-processor communications (e.g., PCIe).

[0055] The compute sled 800 also includes a communication circuit 830. The illustrative communication circuit 830 includes a network interface controller (NIC) 832, which may also be referred to as a host fabric interface (HFI). The NIC 832 may be embodied as, or otherwise include, any type of integrated circuit, discrete circuits, controller chips, chipsets, add-in-boards, daughtercards, network interface cards, or other devices that may be used by the compute sled 800 to connect with another compute device (e.g., with other sleds 400). In some embodiments, the NIC 832 may be embodied as part of a system-on-a-chip (SoC) that includes one or more processors, or included on a multichip package that also contains one or more processors. In some embodiments, the NIC 832 may include a local processor (not shown) and/or a local memory (not shown) that are both local to the NIC 832. In such embodiments, the local processor of the NIC 832 may be capable of performing one or more of the functions of the processors 820. Additionally or alternatively, in such embodiments, the local memory of the NIC 832 may be integrated into one or more components of the compute sled at the board level, socket level, chip level, and/or other levels. In some examples, a network interface includes a network interface controller or a network interface card. In some examples, a network interface can include one or more of a network interface controller (NIC) 832, a host fabric interface (HFI), a host bus adapter (HBA), network interface connected to a bus or connection (e.g., PCIe, CXL, DDR, and so forth). In some examples, a network interface can be part of a switch or a system -on-chip (SoC).

[0056] The communication circuit 830 is communicatively coupled to an optical data connector 834. The optical data connector 834 is configured to mate with a corresponding optical data connector of the rack 240 when the compute sled 800 is mounted in the rack 240. Illustratively, the optical data connector 834 includes a plurality of optical fibers which lead from a mating surface of the optical data connector 834 to an optical transceiver 836. The optical transceiver 836 is configured to convert incoming optical signals from the rack-side optical data connector to electrical signals and to convert electrical signals to outgoing optical signals to the rack-side optical data connector. Although shown as forming part of the optical data connector 834 in the illustrative embodiment, the optical transceiver 836 may form a portion of the communication circuit 830 in other embodiments.

[0057] In some embodiments, the compute sled 800 may also include an expansion connector 840. In such embodiments, the expansion connector 840 is configured to mate with a corresponding connector of an expansion chassis-less circuit board substrate to provide additional physical resources to the compute sled 800. The additional physical resources may be used, for example, by the processors 820 during operation of the compute sled 800. The expansion chassis-less circuit board substrate may be substantially similar to the chassis-less circuit board substrate 602 discussed above and may include various electrical components mounted thereto. The particular electrical components mounted to the expansion chassis-less circuit board substrate may depend on the intended functionality of the expansion chassis-less circuit board substrate. For example, the expansion chassis-less circuit board substrate may provide additional compute resources, memory resources, and/or storage resources. As such, the additional physical resources of the expansion chassis-less circuit board substrate may include, but is not limited to, processors, memory devices, storage devices, and/or accelerator circuits including, for example, field programmable gate arrays (FPGA), application-specific integrated circuits (ASICs), security co-processors, graphics processing units (GPUs), machine learning circuits, or other specialized processors, controllers, devices, and/or circuits.

[0058] Referring now to FIG. 9, an illustrative embodiment of the compute sled 800 is shown. As shown, the processors 820, communication circuit 830, and optical data connector 834 are mounted to the top side 650 of the chassis-less circuit board substrate 602. Any suitable attachment or mounting technology may be used to mount the physical resources of the compute sled 800 to the chassis-less circuit board substrate 602. For example, the various physical resources may be mounted in corresponding sockets (e.g., a processor socket), holders, or brackets. In some cases, some of the electrical components may be directly mounted to the chassis-less circuit board substrate 602 via soldering or similar techniques.

[0059] As discussed above, the individual processors 820 and communication circuit 830 are mounted to the top side 650 of the chassis-less circuit board substrate 602 such that no two heat-producing, electrical components shadow each other. In the illustrative embodiment, the processors 820 and communication circuit 830 are mounted in corresponding locations on the top side 650 of the chassis-less circuit board substrate 602 such that no two of those physical resources are linearly in-line with others along the direction of the airflow path 608. It should be appreciated that, although the optical data connector 834 is in-line with the communication circuit 830, the optical data connector 834 produces no or nominal heat during operation.

[0060] The memory devices 720 of the compute sled 800 are mounted to the bottom side 750 of the of the chassis-less circuit board substrate 602 as discussed above in regard to the sled 400. Although mounted to the bottom side 750, the memory devices 720 are communicatively coupled to the processors 820 located on the top side 650 via the I/O subsystem 622. Because the chassis-less circuit board substrate 602 is embodied as a double-sided circuit board, the memory devices 720 and the processors 820 may be communicatively coupled by one or more vias, connectors, or other mechanisms extending through the chassis-less circuit board substrate 602. Of course, each processor 820 may be communicatively coupled to a different set of one or more memory devices 720 in some embodiments. Alternatively, in other embodiments, each processor 820 may be communicatively coupled to each memory device 720. In some embodiments, the memory devices 720 may be mounted to one or more memory mezzanines on the bottom side of the chassis-less circuit board substrate 602 and may interconnect with a corresponding processor 820 through a ball-grid array.

[0061] Each of the processors 820 includes a heatsink 850 secured thereto. Due to the mounting of the memory devices 720 to the bottom side 750 of the chassis-less circuit board substrate 602 (as well as the vertical spacing of the sleds 400 in the corresponding rack 240), the top side 650 of the chassis-less circuit board substrate 602 includes additional “free” area or space that facilitates the use of heatsinks 850 having a larger size relative to traditional heatsinks used in typical servers. Additionally, due to the improved thermal cooling characteristics of the chassis-less circuit board substrate 602, none of the processor heatsinks 850 include cooling fans attached thereto. That is, each of the heatsinks 850 is embodied as a fan-less heatsink. In some embodiments, the heat sinks 850 mounted atop the processors 820 may overlap with the heat sink attached to the communication circuit 830 in the direction of the airflow path 608 due to their increased size, as illustratively suggested by FIG. 9.

[0062] Referring now to FIG. 10, in some embodiments, the sled 400 may be embodied as an accelerator sled 1000. The accelerator sled 1000 is configured, to perform specialized compute tasks, such as machine learning, encryption, hashing, or other computational-intensive task. In some embodiments, for example, a compute sled 800 may offload tasks to the accelerator sled 1000 during operation. The accelerator sled 1000 includes various components similar to components of the sled 400 and/or compute sled 800, which have been identified in FIG. 10 using the same reference numbers. The description of such components provided above in regard to FIGS. 6, 7, and 8 apply to the corresponding components of the accelerator sled 1000 and is not repeated herein for clarity of the description of the accelerator sled 1000. [0063] In the illustrative accelerator sled 1000, the physical resources 620 are embodied as accelerator circuits 1020. Although only two accelerator circuits 1020 are shown in FIG. 10, it should be appreciated that the accelerator sled 1000 may include additional accelerator circuits 1020 in other embodiments. For example, as shown in FIG. 11, the accelerator sled 1000 may include four accelerator circuits 1020 in some embodiments. The accelerator circuits 1020 may be embodied as any type of processor, co-processor, compute circuit, or other device capable of performing compute or processing operations. For example, the accelerator circuits 1020 may be embodied as, for example, central processing units, cores, field programmable gate arrays (FPGA), application-specific integrated circuits (ASICs), programmable control logic (PCL), security co-processors, graphics processing units (GPUs), neuromorphic processor units, quantum computers, machine learning circuits, or other specialized processors, controllers, devices, and/or circuits.

[0064] In some embodiments, the accelerator sled 1000 may also include an accelerator-to-accelerator interconnect 1042. Similar to the resource-to-resource interconnect 624 of the sled 600 discussed above, the accelerator-to-accelerator interconnect 1042 may be embodied as any type of communication interconnect capable of facilitating accelerator-to-accelerator communications. In the illustrative embodiment, the accelerator-to-accelerator interconnect 1042 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the accelerator-to-accelerator interconnect 1042 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications. In some embodiments, the accelerator circuits 1020 may be daisy-chained with a primary accelerator circuit 1020 connected to the NIC 832 and memory 720 through the I/O subsystem 622 and a secondary accelerator circuit 1020 connected to the NIC 832 and memory 720 through a primary accelerator circuit 1020.

[0065] Referring now to FIG. 11, an illustrative embodiment of the accelerator sled 1000 is shown. As discussed above, the accelerator circuits 1020, communication circuit 830, and optical data connector 834 are mounted to the top side 650 of the chassis-less circuit board substrate 602. Again, the individual accelerator circuits 1020 and communication circuit 830 are mounted to the top side 650 of the chassis-less circuit board substrate 602 such that no two heat-producing, electrical components shadow each other as discussed above. The memory devices 720 of the accelerator sled 1000 are mounted to the bottom side 750 of the of the chassis-less circuit board substrate 602 as discussed above in regard to the sled 600. Although mounted to the bottom side 750, the memory devices 720 are communicatively coupled to the accelerator circuits 1020 located on the top side 650 via the I/O subsystem 622 (e.g., through vias). Further, each of the accelerator circuits 1020 may include a heatsink 1070 that is larger than a traditional heatsink used in a server. As discussed above with reference to the heatsinks 870, the heatsinks 1070 may be larger than traditional heatsinks because of the “free” area provided by the memory resources 720 being located on the botom side 750 of the chassis-less circuit board substrate 602 rather than on the top side 650.

[0066] Referring now to FIG. 12, in some embodiments, the sled 400 may be embodied as a storage sled 1200. The storage sled 1200 is configured, to store data in a data storage 1250 local to the storage sled 1200. For example, during operation, a compute sled 800 or an accelerator sled 1000 may store and retrieve data from the data storage 1250 of the storage sled 1200. The storage sled 1200 includes various components similar to components of the sled 400 and/or the compute sled 800, which have been identified in FIG. 12 using the same reference numbers. The description of such components provided above with regard to FIGS. 6, 7, and 8 apply to the corresponding components of the storage sled 1200 and is not repeated herein for clarity of the description of the storage sled 1200.

[0067] In the illustrative storage sled 1200, the physical resources 620 are embodied as storage controllers 1220. Although only two storage controllers 1220 are shown in FIG. 12, it should be appreciated that the storage sled 1200 may include additional storage controllers 1220 in other embodiments. The storage controllers 1220 may be embodied as any type of processor, controller, or control circuit capable of controlling the storage and retrieval of data into the data storage 1250 based on requests received via the communication circuit 830. In the illustrative embodiment, the storage controllers 1220 are embodied as relatively low-power processors or controllers. For example, in some embodiments, the storage controllers 1220 may be configured to operate at a power rating of about 75 wats.

[0068] In some embodiments, the storage sled 1200 may also include a controller-to-controller interconnect 1242. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the controller-to-controller interconnect 1242 may be embodied as any type of communication interconnect capable of facilitating controller-to-controller communications. In the illustrative embodiment, the controller-to-controller interconnect 1242 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the controller-to-controller interconnect 1242 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point- to-point interconnect dedicated to processor-to-processor communications.

[0069] Referring now to FIG. 13, an illustrative embodiment of the storage sled 1200 is shown. In the illustrative embodiment, the data storage 1250 is embodied as, or otherwise includes, a storage cage 1252 configured to house one or more solid state drives (SSDs) 1254. To do so, the storage cage 1252 includes a number of mounting slots 1256, each of which is configured to receive a corresponding solid state drive 1254. Each of the mounting slots 1256 includes a number of drive guides 1258 that cooperate to define an access opening 1260 of the corresponding mounting slot 1256. The storage cage 1252 is secured to the chassis-less circuit board substrate 602 such that the access openings face away from (e.g., toward the front of) the chassis-less circuit board substrate 602. As such, solid state drives 1254 are accessible while the storage sled 1200 is mounted in a corresponding rack 204. For example, a solid state drive 1254 may be swapped out of a rack 240 (e.g., via a robot) while the storage sled 1200 remains mounted in the corresponding rack 240.

[0070] The storage cage 1252 illustratively includes sixteen mounting slots 1256 and is capable of mounting and storing sixteen solid state drives 1254. Of course, the storage cage 1252 may be configured to store additional or fewer solid state drives 1254 in other embodiments. Additionally, in the illustrative embodiment, the solid state drivers are mounted vertically in the storage cage 1252, but may be mounted in the storage cage 1252 in a different orientation in other embodiments. Each solid state drive 1254 may be embodied as any type of data storage device capable of storing long term data. To do so, the solid state drives 1254 may include volatile and non-volatile memory devices discussed above.

[0071] As shown in FIG. 13, the storage controllers 1220, the communication circuit 830, and the optical data connector 834 are illustratively mounted to the top side 650 of the chassis-less circuit board substrate 602. Again, as discussed above, any suitable attachment or mounting technology may be used to mount the electrical components of the storage sled 1200 to the chassis-less circuit board substrate 602 including, for example, sockets (e.g., a processor socket), holders, brackets, soldered connections, and/or other mounting or securing techniques.

[0072] As discussed above, the individual storage controllers 1220 and the communication circuit 830 are mounted to the top side 650 of the chassis-less circuit board substrate 602 such that no two heatproducing, electrical components shadow each other. For example, the storage controllers 1220 and the communication circuit 830 are mounted in corresponding locations on the top side 650 of the chassis-less circuit board substrate 602 such that no two of those electrical components are linearly in-line with each other along the direction of the airflow path 608.

[0073] The memory devices 720 of the storage sled 1200 are mounted to the bottom side 750 of the of the chassis-less circuit board substrate 602 as discussed above in regard to the sled 400. Although mounted to the bottom side 750, the memory devices 720 are communicatively coupled to the storage controllers 1220 located on the top side 650 via the I/O subsystem 622. Again, because the chassis-less circuit board substrate 602 is embodied as a double-sided circuit board, the memory devices 720 and the storage controllers 1220 may be communicatively coupled by one or more vias, connectors, or other mechanisms extending through the chassis-less circuit board substrate 602. Each of the storage controllers 1220 includes a heatsink 1270 secured thereto. As discussed above, due to the improved thermal cooling characteristics of the chassis-less circuit board substrate 602 of the storage sled 1200, none of the heatsinks 1270 include cooling fans attached thereto. That is, each of the heatsinks 1270 is embodied as a fan-less heatsink.

[0074] Referring now to FIG. 14, in some embodiments, the sled 400 may be embodied as a memory sled 1400. The storage sled 1400 is optimized, or otherwise configured, to provide other sleds 400 (e.g., compute sleds 800, accelerator sleds 1000, etc.) with access to a pool of memory (e.g., in two or more sets 1430, 1432 of memory devices 720) local to the memory sled 1200. For example, during operation, a compute sled 800 or an accelerator sled 1000 may remotely write to and/or read from one or more of the memory sets 1430, 1432 of the memory sled 1200 using a logical address space that maps to physical addresses in the memory sets 1430, 1432. The memory sled 1400 includes various components similar to components of the sled 400 and/or the compute sled 800, which have been identified in FIG. 14 using the same reference numbers. The description of such components provided above in regard to FIGS. 6, 7, and 8 apply to the corresponding components of the memory sled 1400 and is not repeated herein for clarity of the description of the memory sled 1400. [0075] In the illustrative memory sled 1400, the physical resources 620 are embodied as memory controllers 1420. Although only two memory controllers 1420 are shown in FIG. 14, it should be appreciated that the memory sled 1400 may include additional memory controllers 1420 in other embodiments. The memory controllers 1420 may be embodied as any type of processor, controller, or control circuit capable of controlling the writing and reading of data into the memory sets 1430, 1432 based on requests received via the communication circuit 830. In the illustrative embodiment, each memory controller 1420 is connected to a corresponding memory set 1430, 1432 to write to and read from memory devices 720 within the corresponding memory set 1430, 1432 and enforce any permissions (e.g., read, write, etc.) associated with sled 400 that has sent a request to the memory sled 1400 to perform a memory access operation (e.g., read or write).

[0076] In some embodiments, the memory sled 1400 may also include a controller-to-controller interconnect 1442. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the controller-to-controller interconnect 1442 may be embodied as any type of communication interconnect capable of facilitating controller-to-controller communications. In the illustrative embodiment, the controller-to-controller interconnect 1442 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the controller-to-controller interconnect 1442 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point- to-point interconnect dedicated to processor-to-processor communications. As such, in some embodiments, a memory controller 1420 may access, through the controller-to-controller interconnect 1442, memory that is within the memory set 1432 associated with another memory controller 1420. In some embodiments, a scalable memory controller is made of multiple smaller memory controllers, referred to herein as “chiplets”, on a memory sled (e.g., the memory sled 1400). The chiplets may be interconnected (e.g., using EMIB (Embedded Multi-Die Interconnect Bridge)). The combined chiplet memory controller may scale up to a relatively large number of memory controllers and I/O ports, (e.g., up to 16 memory channels). In some embodiments, the memory controllers 1420 may implement a memory interleave (e.g., one memory address is mapped to the memory set 1430, the next memory address is mapped to the memory set 1432, and the third address is mapped to the memory set 1430, etc.). The interleaving may be managed within the memory controllers 1420, or from CPU sockets (e.g., of the compute sled 800) across network links to the memory sets 1430, 1432, and may improve the latency associated with performing memory access operations as compared to accessing contiguous memory addresses from the same memory device.

[0077] Further, in some embodiments, the memory sled 1400 may be connected to one or more other sleds 400 (e.g., in the same rack 240 or an adjacent rack 240) through a waveguide, using the waveguide connector 1480. In the illustrative embodiment, the waveguides are 64 millimeter waveguides that provide 16 Rx (e.g., receive) lanes and 16 Tx (e.g., transmit) lanes. Each lane, in the illustrative embodiment, is either 16 GHz or 32 GHz. In other embodiments, the frequencies may be different. Using a waveguide may provide high throughput access to the memory pool (e.g., the memory sets 1430, 1432) to another sled (e.g., a sled 400 in the same rack 240 or an adjacent rack 240 as the memory sled 1400) without adding to the load on the optical data connector 834. [0078] Referring now to FIG. 15, a system for executing one or more workloads (e.g., applications) may be implemented in accordance with the data center 100. In the illustrative embodiment, the system 1510 includes an orchestrator server 1520, which may be embodied as a managed node comprising a compute device (e.g., a processor 820 on a compute sled 800) executing management software (e.g., a cloud operating environment, such as OpenStack) that is communicatively coupled to multiple sleds 400 including a large number of compute sleds 1530 (e.g., each similar to the compute sled 800), memory sleds 1540 (e.g., each similar to the memory sled 1400), accelerator sleds 1550 (e.g., each similar to the memory sled 1000), and storage sleds 1560 (e.g., each similar to the storage sled 1200). One or more of the sleds 1530, 1540, 1550, 1560 may be grouped into a managed node 1570, such as by the orchestrator server 1520, to collectively perform a workload (e.g., an application 1532 executed in a virtual machine or in a container). The managed node 1570 may be embodied as an assembly of physical resources 620, such as processors 820, memory resources 720, accelerator circuits 1020, or data storage 1250, from the same or different sleds 400. Further, the managed node may be established, defined, or “spun up” by the orchestrator server 1520 at the time a workload is to be assigned to the managed node or at any other time, and may exist regardless of whether any workloads are presently assigned to the managed node. In the illustrative embodiment, the orchestrator server 1520 may selectively allocate and/or deallocate physical resources 620 from the sleds 400 and/or add or remove one or more sleds 400 from the managed node 1570 as a function of quality of service (QoS) targets (e.g., a target throughput, a target latency, a target number instructions per second, etc.) associated with a service level agreement for the workload (e.g., the application 1532). In doing so, the orchestrator server 1520 may receive telemetry data indicative of performance conditions (e.g., throughput, latency, instructions per second, etc.) in each sled 400 of the managed node 1570 and compare the telemetry data to the quality of service targets to determine whether the quality of service targets are being satisfied. The orchestrator server 1520 may additionally determine whether one or more physical resources may be deallocated from the managed node 1570 while still satisfying the QoS targets, thereby freeing up those physical resources for use in another managed node (e.g., to execute a different workload). Alternatively, if the QoS targets are not presently satisfied, the orchestrator server 1520 may determine to dynamically allocate additional physical resources to assist in the execution of the workload (e.g., the application 1532) while the workload is executing. Similarly, the orchestrator server 1520 may determine to dynamically deallocate physical resources from a managed node if the orchestrator server 1520 determines that deallocating the physical resource would result in QoS targets still being met.

[0079] Additionally, in some embodiments, the orchestrator server 1520 may identify trends in the resource utilization of the workload (e.g., the application 1532), such as by identifying phases of execution (e.g., time periods in which different operations, each having different resource utilizations characteristics, are performed) of the workload (e.g., the application 1532) and pre-emptively identifying available resources in the data center 100 and allocating them to the managed node 1570 (e.g., within a predefined time period of the associated phase beginning). In some embodiments, the orchestrator server 1520 may model performance based on various latencies and a distribution scheme to place workloads among compute sleds and other resources (e.g., accelerator sleds, memory sleds, storage sleds) in the data center 100. For example, the orchestrator server 1520 may utilize a model that accounts for the performance of resources on the sleds 400 (e.g., FPGA performance, memory access latency, etc.) and the performance (e.g., congestion, latency, bandwidth) of the path through the network to the resource (e.g., FPGA). As such, the orchestrator server 1520 may determine which resource(s) should be used with which workloads based on the total latency associated with each potential resource available in the data center 100 (e.g., the latency associated with the performance of the resource itself in addition to the latency associated with the path through the network between the compute sled executing the workload and the sled 400 on which the resource is located).

[0080] In some embodiments, the orchestrator server 1520 may generate a map of heat generation in the data center 100 using telemetry data (e.g., temperatures, fan speeds, etc.) reported from the sleds 400 and allocate resources to managed nodes as a function of the map of heat generation and predicted heat generation associated with different workloads, to maintain a target temperature and heat distribution in the data center 100. Additionally or alternatively, in some embodiments, the orchestrator server 1520 may organize received telemetry data into a hierarchical model that is indicative of a relationship between the managed nodes (e.g., a spatial relationship such as the physical locations of the resources of the managed nodes within the data center 100 and/or a functional relationship, such as groupings of the managed nodes by the customers the managed nodes provide services for, the types of functions typically performed by the managed nodes, managed nodes that typically share or exchange workloads among each other, etc.). Based on differences in the physical locations and resources in the managed nodes, a given workload may exhibit different resource utilizations (e.g., cause a different internal temperature, use a different percentage of processor or memory capacity) across the resources of different managed nodes. The orchestrator server 1520 may determine the differences based on the telemetry data stored in the hierarchical model and factor the differences into a prediction of future resource utilization of a workload if the workload is reassigned from one managed node to another managed node, to accurately balance resource utilization in the data center 100. In some embodiments, the orchestrator server 1520 may identify patterns in resource utilization phases of the workloads and use the patterns to predict future resource utilization of the workloads.

[0081] To reduce the computational load on the orchestrator server 1520 and the data transfer load on the network, in some embodiments, the orchestrator server 1520 may send self-test information to the sleds 400 to enable each sled 400 to locally (e.g., on the sled 400) determine whether telemetry data generated by the sled 400 satisfies one or more conditions (e.g., an available capacity that satisfies a predefined threshold, a temperature that satisfies a predefined threshold, etc.). Each sled 400 may then report back a simplified result (e.g., yes or no) to the orchestrator server 1520, which the orchestrator server 1520 may utilize in determining the allocation of resources to managed nodes.

MEMORY INTERFACE

[0082] Various embodiments provide for writing memory access commands (e.g., read, write, read- modify-write) to a region of memory rather than to device space (e.g., PCIe or some other input output (IO) interface). For some memory pooling models such as page-based memory pooling models, a data accessor (e.g., application, virtual machine, container, processor, circuitry, or accelerator) can write a direct memory access (DMA) command (e.g., source address, destination address, length) to a region of memory rather than issue a network access command (e.g., request packet transmission) to access data in a remote memory pool over a fabric or network. Some embodiments can be used in latency sensitive or latency insensitive environments. For example, some embodiments can be used for memory accesses at page granularity.

[0083] Various embodiments provide a DMA engine that is controlled or commanded by writing to a memory region in system memory address space. A command to access a memory pooling (e.g., “read from pool”, “write to pool”, etc.) could be provided to the DMA engine via a write to the memory region. For commands, the DMA engine could claim or be assigned a region of platform address space, as with other agents or devices connected to a memory interface. The DMA engine could be connected to a host system using a memory interface (e.g., CXL, DDR, UPI, etc.) and registered as a memory device with an amount of memory for a memory command portal. In some examples, platform firmware can allocate part of system memory to this command portal.

[0084] In some examples, writes to the memory region can be detected by polling by the DMA engine or interrupts issued to the DMA engine. The memory region can be allocated by an operating system or driver, for example, for a requester to write data access commands to the memory region to control operation of a DMA engine. System software could interact with this memory region as generally available to be written-to or read-from by software or devices other than a memory access requester or the DMA engine. However, in some examples, writes to the memory region are available to the DMA engine and are not copied to any cache. The memory region may be mapped as write -through or using cache invalidate instructions to force data copied from the memory region to a cache to be evicted. In some cases, reads from the memory region can be ignored. In some examples, the memory region may not be used to receive replies from a memory pool and the memory fabric may not be stalled to wait for replies from the remote pool.

[0085] Various embodiments provide for copying data to or from a local memory device using a memory fabric interface instead of using a device interface (e.g., PCIe). Use of an IO interface such as PCIe can introduce higher latency than use of a memory interface for writes to memory or reads from memory because IO interfaces perform semantic conversion of an IO command to use of memory interfaces (e.g., conversion to address bus line communication semantics, conversion to data command line communication semantics, and conversion to buffer access semantics) whereas a command sent using a memory interface can be provided without format conversion.

[0086] For a data write command involving a source memory address associated with local memory and a destination memory address associated with a memory pool node, the DMA engine can receive a command in a memory region allocated to receive data access requests. The DMA engine can use a memory interface to access data associated with the source memory address in the node memory using a memory interface and transmit the data in a packet to the remote memory pool node using a network interface. For example, a memory interface can be at least one of: CXL, DDR, UPI, and others). In some example, a memory interface can provide for lower latency reads or writes to a local memory device than that provided by a block-based IO device (e.g., PCIe, or CXL.io). The network interface can form a packet and transmit the packet with the data to the remote memory pool node.

[0087] For a data read command involving a destination memory address associated with local memory and a source memory address associated with a remote memory pool node, the DMA engine can receive a command in a memory region allocated to receive data access requests. The DMA engine can use a network interface to form a packet and transmit the packet with a data access request to the remote memory pool node. The network interface can decapsulate the data from the packet and the DMA engine can use a memory interface to write received data to the local memory.

[0088] FIG. 16 depicts an example of memory pools accessible by a computing node. Compute node 1600 can include a network or fabric interface, compute resources (e.g., CPUs or accelerators), and memory resources (e.g., memory, storage, or cache). For example, compute node 1600 can execute workloads or applications. For example, compute node 1600 can be implemented as a server, rack of servers, computing platform, or others. In some examples, a host node can include one or more of: a core, graphics processing unit (GPU), field programmable gate array (FPGA), or application specific integrated circuit (ASIC). In some examples, a core can be sold or designed by Intel®, ARM®, AMD®, Qualcomm®, IBM®, Texas Instruments®, among others. Any processor can execute an operating system, driver, applications, and/or a virtualized execution environment (VEE) (e.g., virtual machine or container). In some examples, an operating system (OS) can be Linux®, Windows®, FreeBSD®, Android®, MacOS®, iOS®, or any other operating system.

[0089] Memory 1608 can include one or more of: one or more registers, one or more cache devices (e.g., level 1 cache (LI), level 2 cache (L2), level 3 cache (L3), lower level cache (LLC)), volatile memory device, non-volatile memory device, or persistent memory device. For example, memory 1608 can include static random access memory (SRAM) memory technology or memory technology consistent with high bandwidth memory (HBM), or double data rate (DDR), among others. Memory 1608 can be connected to network interface 1604 using a high speed interface (e.g., DDR, CXL (e.g., Compute Express Link Specification revision 2.0, version 0.9 (2020), as well as earlier versions, revisions or variations thereof), Peripheral Component Interconnect express (PCIe) (e.g., PCI Express Base Specification 1.0 (2002), as well as earlier versions, revisions or variations thereof).

[0090] Compute node 1600 can be coupled to one or more of memory pool nodes 1650-0 to 1650-M- I using a network or fabric. Any memory pool node 1650 can include a network or fabric interface 1653, compute resources (e.g., CPUs 1654, GPUs, or accelerators), and memory pool resources 1656 (e.g., memory, storage, or cache). Although this example shows merely one compute node coupled to multiple memory pool nodes, multiple compute nodes can be coupled to multiple memory pool nodes. For example, any of memory pool nodes 1650-0 to 1650-M-l can be coupled to network interface 1604 using a network interface and one or more switches or routers.

[0091] In some examples of a memory pool node, a CPU 1654 can be part of a SoC and network interface 1652 can access a memory device using an uncore without involving the SoC or CPU 1654. In some examples of a memory pool node, network interface 1652 can could interact with a memory controller to perform read or write memory devices in a memory pool 1656.

[0092] A memory access request (e.g., read, write, read-modify-write) can be issued by an application, virtualized execution environment (VEE) (e.g., virtual machine or container), operating system (OS) or other software executed by CPUs 1606 or other device such as an accelerator device. For example, an OS can determine that content in a page of addressable memory has not been accessed from local memory 1608 within an amount of time by any copy operation of content in the page to a cache or a read operation and determine that the page of content is not likely to be accessed within an amount of time, and the OS can cause content in the page to be evicted to a memory pool node to free space in memory 1608 for other uses. Such page can be considered “cold.” Example memory eviction schemes include least recently used (LRU), Time aware Least Recently Used (TLRU) (e.g., Bilal, Muhammad; etal. (2017). "Time Aware Least Recent Used (TLRU) Cache Management Policy in ICN". IEEE 16th International Conference on Advanced Communication Technology (ICACT): 528-532), set associativity, or others.

[0093] System software such as an OS could manage local memory 1608 such that its content is frequently accessed and blocks in remote memory (e.g., in any memory pool node 1650-0 to 1650-M-l) are infrequently accessed. In this approach, system software moves data between the remote memory pool and the local memory with the goal of keeping “hot” data in the local memory and “cold” data in the remote memory. Some embodiments manage memory content at a page granularity, although larger or smaller regions of memory can be marked as cold or hot. Hardware, software, or some combination of the two may manage data content in local or remote memory.

[0094] Contents of memory 1612 could be managed in any number of manners. For example, memory 1612 could be managed as a cache with appropriate pre-fetching, victimization, eviction, and other cache content policies. Memory 1612 could be managed as a tier of memory that can be directly mapped into the system and managed by system software (e.g., OS, hypervisor (e.g., Linux, VMware ESX, Windows Hyper- V), orchestrator (e.g., Kubemetes, OpenStack, Slurm (High Performance Computing (HPC)), Open Source NFV Management and Orchestration (MANO) from European Telecommunications Standards Institute (ETSI)’s Open Source Mano (OSM) group), pod manager, or traffic manager on a same or different host node, fabric manager (e.g., CXL fabric manager). Memory 1608 could be managed as a pre-fetch buffer driven by policies running in network interface 1604 (e.g., smartNIC) or local CPUs 1606.

[0095] A page can be 4096 bytes, but larger or smaller sizes or regions of memory can be monitored to determine whether to evict the content to a memory pool. However, in some examples, the decision to evict the content to a memory pool may be faulty as an application or device can request access to the content. In some cases, compute node 1600 requests the data from a memory pool node to service a page fault (e.g., requested data is not available in local memory 1608) and such request can introduce latency between when the data is requested to be accessed and when the data is available to process.

[0096] In some examples, DMA engine 1602 can perform a DMA operation within local memory 1608. DMA allows a CPU or core to not manage a copy operation when sending or receiving data to or from the system memory, the CPU or core can be available to perform other operations. Without DMA, when the CPU or core is using programmed input/output, the CPU or core is typically occupied for the entire duration of a read or write operation and is unavailable to perform other work. With DMA, the CPU or core can, for example, initiate a data transfer, and then perform other operations while the data transfer is in progress. The CPU or core can receive an interrupt or other notification from a DMA controller when the data transfer is finished.

[0097] In some examples, DMA engine 1602 can provide a network interface to a fabric or network to communicate at least with any of memory pool nodes 1650-0 to 1650-M-l. In cases where a source address corresponds to a memory region in any of memory pools 1650-0 to 1650-M-l, but the destination address corresponds to a memory region in local memory 1608, DMA engine 1602 can utilize network interface 1604 to request a copy of content from one or more of memory pools 1650-0 to 1650-M-l by transmitting packets where the packets can convey a memory request with a read request and the source address and length. DMA engine 1602 can copy the received content into a specified destination address in local memory 1608. Some examples can use RDMA or NVMe-oF protocols to access content from remote memory.

[0098] If a source address or destination address is in any of memory pools 1650-0 to 1650-M-l, DMA engine 1602 can generate one or more packets with a memory request. Where a source address corresponds to a memory device in any of memory pools 1650-0 to 1650-M-l, the packets can convey a memory request with a read request and the source address and length. Where a destination address corresponds to a memory device in any of memory pools 1650-0 to 1650-M-l, the packets can convey a memory request with a write request and the destination address and length. A memory pool can respond to a read request by providing the requested content, where the request is authorized. A memory pool can respond to a write request by providing an acknowledgement that the write request was completed, where the request is authorized.

[0099] In cases where the source and destination addresses correspond to memory regions in any of memory pools 1650-0 to 1650-M-l, DMA engine 1602 can utilize network interface 1604 to request a memory pool to perform a copy operation of content to another memory pool.

[00100] If a source address or destination address is a virtual address, DMA engine 1602 can perform virtual address translation to generate a physical address corresponding to the virtual address. Some embodiments utilize virtual address translation in a DMA engine for remote data access without the requester managing network buffers. For example, a requester can access local buffers in local memory 1608 and not manage network connections, setup or shutdown. DMA engine 1602 can manage virtual addresses and translation

[00101] Various embodiments of DMA engine 1602 can be accessed by a requester (e.g., application, virtual machine, container, operating system, driver, and so forth) using DMA semantics. In some examples, a DMA data transfer request can be specified using the following format:

Data_transfer(source address, destination address, length).

[00102] DMA engine 1602 could inspect address information (e.g., a physical or virtual address, address space identifier, etc.) in the DMA data transfer request and use that information to determine where the source or destination data is located (e.g., in local memory 1608 or in remote pool memory of some memory pool attached to the overall system via one or more switches or routers). When the address information indicates source or destination is remote, DMA engine 1602 can use network interface 1604 to carry messages to the remote memory pool to facilitate a read or write of pool memory. The messaging can support read or write semantics with appropriate reliability guarantees such as packet receipt Acknowledgements or negative-acknowledgement (NACK), reliable transport protocol (RTP), Cyclic Redundancy Check (CRC), checksums, re-transmit, packet sequence numbering, remote direct memory access (RDMA), UDP, and so forth.

[00103] DMA is a technology that allows an input/output (I/O) device to bypass a central processing unit (CPU) or core, and to send or receive data directly to or from a system memory. Because DMA allows the CPU or core to not manage a copy operation when sending or receiving data to or from the system memory, the CPU or core can be available to perform other operations. Without DMA, when the CPU or core is using programmed input/output, the CPU or core is typically occupied for the entire duration of a read or write operation and is unavailable to perform other work. With DMA, the CPU or core can, for example, initiate a data transfer, and then perform other operations while the data transfer is in progress. The CPU or core can receive an interrupt from a DMA controller when the data transfer is finished.

[00104] Some examples of DMA engine 1602 provide data mover and transformation operations. For example, DMA engine 1602 can validate CRC or checksum values in connection wtih storage and networking applications. For example, DMA engine 1602 can perform memory compare and delta generation or merge to support VM or container migration, VM or container check-pointing (e.g., to revert a VM or container to a previous state) and software managed memory deduplication usages.

[00105] Some examples of DMA engine 1602 are part of an Infrastructure Processing Unit (IPU). An IPU can include a SmartNIC with one or more programmable or fixed function processors to perform offload of operations that could have been performed by a CPU. The IPU can include one or more memory devices. In some examples, the IPU can perform virtual switch operations, manage storage transactions (e.g., compression, cryptography, virtualization), and manage operations performed on other IPUs, servers, or devices.

[00106] In some examples, DMA engine 1602 and/or network interface 1604 can be formed in a system on a chip (SOC) with one or more of CPUs 1606 and local memory 1608. In some examples, DMA engine 1602 and/or network interface 1604 can be coupled to one or more of CPUs 1606 and local memory 1608 using a device interface (e.g., DDR, CXU, PCIe). In some examples, network interface 1604 can directly access local memory 1608 without intervention by CPUs 1606 to access data.

[00107] In some examples, to communicate with any of memory pools 1650-0 to 1650-M-l, network interface 1604 can utilize a remote direct memory access (RDMA) protocol with gets and puts or reads or writes. Examples of RDMA protocols include InfiniBand, Internet Wide Area RDMA Protocol (iWARP), quick UDP Internet Connections (QUIC), RDMA over Converged Ethernet (RoCE). In some examples, RDMA-enabled network interface is managed by requester software by management of its network construct, establishing a connection and managing network events. Some embodiments allow a requester to merely issue a DMA request and DMA engine 1602 and network interface 1604 handle setting up a connection, connection management, and RDMA setup with a target memory node. Accordingly, a requester may not need to initiate connection set up or perform a packet transmission but can merely issue a DMA request. The requester can specify source and destination address and length to DMA engine 1602 through a driver or library.

[00108] For example, for a put or get operation, network interface 1604 can be RDMA-enabled to setup remote RDMA operations including send queues and receive queues. RDMA can involve direct writes or reads to copy content of buffers across a connection without the operating system managing the copies. A network interface or other interface to a connection can implement a direct memory access engine and create a channel from its RDMA engine though a bus to application memory. A send queue and receive queue can be used to transfer work requests and are referred to as a Queue Pair (QP). A requester can place work request instructions on its work queues that tells the interface contents of what buffers to send to or receive content from. A work request can include an identifier (e.g., pointer or memory address of abuffer). For example, a work request placed on a send queue (SQ) can include an identifier of a message or content in a buffer (e.g., app buffer) to be sent. By contrast, an identifier in a work request in a Receive Queue (RQ) can include a pointer to a buffer (e.g., app buffer) where content of an incoming message can be stored. An RQ can be used to receive an RDMA-based command or RDMA -based response. A Completion Queue (CQ) can be used to notify when the instructions placed on the work queues have been completed.

[00109] For a get or read operation, network interface 1604 can send one or more packets with a get or read operation, with source address (e.g., translated to a physical address or not translated to a physical address) and length, to a target memory node. Network interface 1604 can copy the received data from a memory node to a destination address in local memory 1608.

[00110] For example, a “pull” model can be used whereby a compute node can send “put from address x” to a memory pool and memory pool responds with “get from address x” to compute node and compute node responds with data to memory pool.

[00111] DMA engine 1602 can utilize an input/output (I/O) device interface 1610 to communicate with CPUs 1606 or core. DMA engine 1602 can utilize a memory interface 1612 to communicate with local memory 1608. In some examples, the memory interface provides a lower latency and higher speed data write or read involving respective storage to or reading from local memory 1608.

[00112] In some examples, IO interface can support control path interactions with DMA engine 1602 to initialize and configure DMA engine 1602. Initializing and configuring DMA engine 1602 can include configuring DMA engine 1602 to read commands from a command portal region of a memory as described herein as well as identifying which addresses correspond to local memory 1608 or to devices in memory pool 1650-0 to 1650-M-l. Initializing and configuring DMA engine 1602 may be infrequent operations that are not time critical. System software can configure DMA engine 1602 using techniques such as writing to memory mapped input output (MMIO) registers, and so forth via the IO interface. [00113] In some examples, memory interface 1612 can be utilized to read or write data from or to local memory 1608 in response to a command from CPUs 1606 or a message or packet received over a network medium or fabric from a memory pool node 1650-0 to 1650-M-l. In some examples, a command portal can be provided in a memory address region assigned to DMA engine 1602. To issue a command, one or more of CPUs 1606 writes an N-byte descriptor that specifies a source node, source address, destination node, destination address, and length of data to the memory address region. In some examples, multiple memory address regions can be used such as a memory address region to receive a read or write command involving local memory 1608 as a source or destination and another memory address region to receive reads or writes involving a memory pool as a source or destination.

[00114] In some examples, any compute node or any memory pool node can utilize a DMA engine with a memory interface to a memory device to access data from memory or write data to memory. In some examples, source and destination addresses can be in local memory or remote memory. In a case where source and destination addresses are associated with remote memory, a sender DMA engine can send a request to a remote DMA engine associated with a node (e.g., compute or memory) with a local memory that corresponds to the source destination address to perform a copy operation to the destination address using a memory interface or packet(s).

[00115] High speed interconnects can be used to provide communications between compute node 1600 and any of memory nodes 1650-0 to 1650-M-l such as one or more of: Ethernet (IEEE 802.3), remote direct memory access (RDMA), InfiniBand, Internet Wide Area RDMA Protocol (iWARP), Transmission Control Protocol (TCP), User Datagram Protocol (UDP), quick UDP Internet Connections (QUIC), RDMA over Converged Ethernet (RoCE), Peripheral Component Interconnect express (PCIe), Intel QuickPath Interconnect (QPI), Intel® Ultra Path Interconnect (UPI), Intel On-Chip System Fabric (IOSF), Omnipath, Compute Express Link (CXL), HyperTransport, high-speed fabric, NVLink, Advanced Microcontroller Bus Architecture (AMBA) interconnect, OpenCAPI, Gen-Z, Cache Coherent Interconnect for Accelerators (CCIX), Infinity Fabric (IF), 3GPP Long Term Evolution (LTE) (4G), 3GPP 5G, and variations thereof. In some examples, data can be copied or stored to virtualized storage nodes using protocols such as Non- Volatile Memory Express (NVMe) or NVMe over fabrics (NVMe-oF) (or iSCSI storage command generation). For example, NVMe-oF is described at least in NVM Express, Inc., “NVM Express Over Fabrics,” Revision 1.0, Jun. 5, 2016, and specifications referenced therein and variations and revisions thereof.

[00116] FIG. 17 depicts an example system. Management entity 1720 can set up a connection between network interface 1708 and remote network interface 1752. Management entity 1720 can discover capacity in memory pool node 1750 and other memory pool nodes for use or access by DMA engine 1706. DMA engine 1706 can receive a configuration from management entity 1720 of which memory pool nodes that DMA engine 1706 can access and also receive address-to-memory node translation information, for example, as described earlier. For example, compute node 1700 can discover availability of memory in one or more memory pool nodes by querying management entity 1720. Management entity 1720 can be implemented as an orchestrator for a data center, edge node, server, rack of servers, or racks of servers and other examples are described herein.

[00117] In some examples, processor 1702 can execute driver 1704. Driver 1704 can set up a connection between network interface 1708 used by DMA engine 1706 and remote network interface 1752 used by memory pool 1750. In some examples, DMA engine 1706 and network interface 1708 can manage connection state context for a connection between network interface 1708 and remote network interface 1752. Connection state context can include one or more of: outstanding requests, sequence numbers, acknowledgement (ACK) sequence numbers, telemetry information to identify packet errors, round trip time (RTT), bandwidth used, destination IP addresses, connection port. For example, of UDP over IPv4, connection state context can include at least a destination IP address and ingress port (e.g., port at a destination IP that the packet is sent to).

[00118] FIG. 18 depicts an example of message transmission ordering. Various embodiments permit out-of-order completion of data transfers using DMA transactions whereby a DMA engine can be permitted to perform data copy operations out-of-order of received DMA requests. In some examples, DMA engine can place data into local memory even if a packet is received out-of-order from transmission from its source. In some examples, DMA engine does not perform buffering to reorder packets to an order of transmission sequence. Avoiding packet reordering can reduce buffer sizes. For example, if packets are routed through network according to a best path at a given time (e.g., least congestion through one or more switches or routers), instead of packets following a path of earlier transmitted packets, packets can be received at a destination DMA engine out of order. Unordered transactions can be performed whereby transactions are sent in any order to a DMA engine from a remote memory node.

[00119] In this example, memory node 1800 sends packets 0, 1, and 2 (with data) to compute node 1810 and packet 2 arrives at network interface 1814, followed by packet 0 and packet 1. Packets 0-2 are shown to take different paths to network interface 1814, although they could take the same path or some packets could take the same path through network elements to network interface 1814 whereas some packets take another path through network elements to network interface 1814. According to various embodiments, DMA engine 1812 can store data from received packets in the order of receipt, namely, data from packet 2, followed by data from packet 0 and followed by data from packet 1 instead of buffering data and reordering data prior to storage.

[00120] FIG. 19A depicts an example of commanding a DMA engine. In this example, a requester executed by a CPU can write a command to a memory address range that corresponds to DMA portal 1904 that is used to receive commands that are to be performed by DMA engine 1900. In this example, a requester can write a command “write page X to page P:Y” where page X represents a memory address in local memory address and page P:Y represents a destination memory address in a remote memory pool node. DMA engine 1900 can read the command from DMA portal 1904 using the memory interface. In some examples, data to be read and copied to a memory pool is written to DMA portal 1904 by the requester with the command. DMA engine 1900 can packetize the write request with data read from page X and send the write request with data to a destination memory pool to send a write command.

[00121] In some examples, a requester can write a DMA request to DMA portal 1904 of a format (source address, destination address, length). DMA engine 1900 can determine if the source or destination addresses correspond to local or remote memory devices. If the source address is in memory 1902, DMA engine 1900 can copy the data from memory 1902 via a memory interface and can cause a network interface to send the data in one or more packets with a write command to the appropriate destination memory device. If the source address is in a remote memory pool node, DMA engine 1900 can cause a network interface to send the read command in one or more packets to the appropriate destination memory device and copy data received in packets from the remote memory pool node to memory 1902 via a memory interface.

[00122] FIG. 19B depicts an example of a read operation. In this example, a requester executed by a CPU can write a command to a memory address range that corresponds to DMA portal 1904 that is used to receive commands that are to be performed by DMA engine 1900. In this example, a requester can write a command “write page P:Y to page X.” DMA engine 1900 can read the command from DMA portal 1904 using a memory interface. DMA engine 1900 can cause a network interface to send the read request in one or more packets to a destination memory pool. DMA engine 1900 can copy data received in packets from the remote memory pool, in response to the read request, to memory 1902 via a memory interface.

[00123] FIG. 20 depicts an example process. The process can be performed by a platform with a DMA engine and network interface in some examples. At 2002, a DMA engine of a compute node can be configured to recognize memory addresses that correspond to memory devices in a local memory or in a remote memory pool. For example, a remote memory device can be coupled to a network interface that issues a memory access request a network interface and one or more switches or routers. At 2004, a memory device can be configured to assign a range of memory addresses for receipt of commands to be performed by the DMA engine. The memory device can be a system memory device accessible to processor-executed software and the DMA engine. At 2006, a determination is made if a command is written to the command portal. If a new command is detected as written to the command portal, then the process can continue to 2008. If a new command is not detected as written to the command portal, then the process can return to 2006.

[00124] At 2008, a determination is made as to whether a source or destination address associated with the command are associated with local or remote memory devices. In some examples, configuration of the DMA engine can indicate whether source addresses and destination addresses correspond to local or remote memory devices. At 2010, the DMA engine can utilize a memory interface to read data from a local memory device or write data to the local memory device. For example, if the source address corresponds to the local memory device and the destination address corresponds to a remote memory device, the DMA engine can utilize the memory interface to read the data and packetize the data to be sent using a network interface in one or more packets to the remote memory device. For example, if the source address corresponds to the remote memory device and the destination address corresponds to a local memory device, the DMA engine can packetize a read request to be sent using a network interface in one or more packets to the remote memory device and utilize the memory interface to copy received data from the remote memory device to the local memory device.

[00125] Accordingly, various embodiments provide a DMA engine or network interface with a memory interface to be utilized to read commands, read data from local memory or write data to local memory. The commands and data can be accessed more rapidly using a memory interface than a device interface and latency of data transfer can be reduced.

GUIDED PREFETCH FOR FUNCTION AS A SERVICE (FAAS) DEPLOYMENTS OPERATING IN DISAGGREGATED MEMORY ENVIRONMENTS

[00126] Various embodiments include a network interface or DMA engine that utilizes a prefetch engine to support prefetching of appropriate information (e.g., data, metadata, and/or executable instructions) from memory based on which function, in a FaaS workload, is currently executing or will be executed.

[00127] FIG. 21 depicts an example system. In this example, compute node 2100 is communicatively coupled to memory pool nodes 2150-0 to 2150-M using network interface 2102 and network interfaces 2152-0 to 2152-M. Instead of or in addition to use of network interfaces, a DMA engine with network interface capability as described herein can be utilized. In compute node 2100, network interface 2102 can utilize a prefetch engine (PFE) 2104. PFE 2104 can be setup, in a function as a service (FaaS) workload or other workload, with information on which remote data or instructions should be moved into local memory 2108. A serverless application or function (e.g., Function as a Service (FaaS)) can use available computing resources at a data center, edge computing device, or fog network device or in disaggregated computing and memory resources. For example, for a serverless application, a cloud service provider dynamically manages allocation and provisioning of servers and a serverless application runs in stateless compute containers that are event-triggered and may last for one invocation. A serverless application can be event- driven, cloud-based application where application development relies on a combination of third-party services, client-side logic and cloud-hosted remote procedure calls. Serverless application can be pay-per- use computing or bare-code where users are charged based on time and computing resources (e.g., CPU, networking, or memory) allocated to run serverless application without associated fees for idle time of computing resources. In some examples, a serverless application or function can be performed by a network infrastructure device (e.g., forwarding element, router, switch, network interface controller) or accelerator, in addition or alternative to use of a server or general purpose computing platform.

[00128] In a FaaS deployment, orchestration (e.g., European Telecommunications Standards Institute (ETSI) Open Source NFV Management and Orchestration (MANO), Kubemetes, OpenStack, Network Function Virtualization (NFV) orchestrator, or others) could set up and configure PFE 2104. This setup could be done at various points of time in the lifetime of a function such as prior to execution to start to retrieve data in cold pages (e.g., less frequently accessed data) for a new function to local memory 2108, following execution of a function to copy data in cold pages for an existing function out of local memory 2108 to a memory pool node.

[00129] PFE 2104 can cause data copy or pre-fetch operations at a variety of granularities. For example, pre-fetching content can include retrieving content and storing the content to memory (e.g., local memory 2108) prior to a request for retrieval of the content issued by a requester or in anticipation that a request for retrieval of the content will be issued, even if the request is never issued. To support memory disaggregation that uses page-based approaches, PFE 2104 could cause complete pages of data to be copied to local memory 2108 from one or more memory nodes or from one or more memory nodes to local memory 2108 or a processor cache. To support cacheline granularity, PFE 2104 might move smaller granularities (e.g., a cacheline or cachelines). PFE 2104 can be implemented as one or more of: field programmable gate array, core, CPU, GPU, application specific integrated circuit (ASIC), or any programmable device or circuitry.

[00130] In some examples, orchestrator 2170 can configure PFE 2104 with function information and data that is to be accessed by each function in a FaaS, an address of the data in any memory device, and a node that includes the memory device. For example, orchestrator 2100 can configure PFE 2104 with the following information.

[00131] Orchestrator 2170 can schedule execution of a function on any compute node in one or more of a server, data center or edge node. Orchestrator 2170 can configure PFE 2104 using a management interface with the information in the table above and indicate to PFE 2104 which function executes on compute node 2100 and which function is to execute after the current function. PFE 2104 can fetch a function (e.g., execution code, operation codes, executable binaries) and/or data to be processed from any memory pool node into a cache or memory of compute node 2100 so that the function and/or data is available in the cache or memory. Delays between commencement of functions in a FaaS or availability of data can be reduced by use of PFE 2104. Orchestrator 2170 can execute in any server, rack, edge node, and so forth.

[00132] In response to a request to prefetch, PFE 2104 can cause network interface 2102 to generate one or more packets with a read request that indicates a source memory address and length of content (e.g., instructions and/or data) and send the one or more packets to any of memory pool node 2150-0 to 2150-M-

1. PFE 2104 can cause network interface 2102 to copy content to a cache or memory of compute node 2100.

[00133] FIG. 22 depicts an example sequence. In this example, a FaaS includes function 0, function 1, followed by function 2. Completion of function 0 causes a prefetcher to prefetch executable instructions and data from respective memory node A and memory node B for function 1. Execution of function 1 utilizes the fetched executable instructions and data. Completion of function 1 causes a prefetcher to prefetch executable instructions and data from respective memory node C and memory node D for function

2. Instructions and data can be stored in memory or cache of a compute node that is to perform function 2. Execution of function 2 can utilize the prefetched executable instructions and data from respective nodes C and D. In other examples, prefetching of executable instructions and data can be performed for more than one function (e.g., two or more). For example, if function 1 is a short lived function, executable instructions and data for functions 2 and 3 (or more) can be prefetched. In some examples, prefetching can occur during execution of a function and before completion of the function. For example, prior to completion of function 0, executable code and data for one or more of functions 1 and 2 can be prefetched into cache or memory of the node that is to execute functions 1 and 2.

[00134] FIG. 23 depicts an example process. The process can be performed by a computing platform with prefetch engine and network interface, in some examples. At 2302, a prefetch engine can be configured with memory addresses and device identifiers that store executable code and data for functions in a sequence of functions. For example, a workload orchestrator can configure the prefetch engine with such information. At 2304, the prefetch engine can prefetch executable code and/or data for one or more functions that are to execute currently or in the future. For example, the orchestrator can indicate which one or more functions are to execute currently and in the future. The prefetch engine can determine which code and/or data to prefetch to cache or memory of a node that is to perform a function. In some examples, if there is insufficient cache or memory space, prefetching occurs to the extent of cache or memory space available. In some examples, regardless of available cache or memory and even if an eviction of data is caused to make space for prefetched code and/or data, the code and/or data are prefetched and stored in cache or memory of a computing node that performs a function.

[00135] At 2306, the prefetched code and/or data are stored into cache or memory of the node that performs the function. The node can include any type of processor that performs the code. The node can be the same or different node that includes the prefetcher. The prefetcher may prefetch code and/or data for another node in some examples.

[00136] FIG. 24 depicts a network interface that can use embodiments or be used by embodiments. In some embodiments, network interface an include capability to access DMA requests from a memory region, perform DMA operations, or pre-fetch content in accordance with embodiments described herein. In some examples, network interface 2400 can be implemented as a network interface controller, network interface card, a host fabric interface (HFI), or host bus adapter (HBA), and such examples can be interchangeable. Network interface 2400 can be coupled to one or more servers using a bus, PCIe, CXL, or DDR. Network interface 2400 may be embodied as part of a system-on-a-chip (SoC) that includes one or more processors, or included on a multichip package that also contains one or more processors.

[00137] Network interface 2400 can include transceiver 2402, processors 2404, transmit queue 2406, receive queue 2408, memory 2410, and bus interface 2412, and DMA engine 2452. Transceiver 2402 can be capable of receiving and transmitting packets in conformance with the applicable protocols such as Ethernet as described in IEEE 802.3, although other protocols may be used. Transceiver 2402 can receive and transmit packets from and to a network via a network medium (not depicted). Transceiver 2402 can include PHY circuitry 2414 and media access control (MAC) circuitry 2416. PHY circuitry 2414 can include encoding and decoding circuitry (not shown) to encode and decode data packets according to applicable physical layer specifications or standards. MAC circuitry 2416 can be configured to perform MAC address filtering on received packets, process MAC headers of received packets by verifying data integrity, remove preambles and padding, and provide packet content for processing by higher layers. MAC circuitry 2416 can be configured to assemble data to be transmitted into packets, that include destination and source addresses along with network control information and error detection hash values.

[00138] Processors 2404 can be any a combination of a: processor, core, graphics processing unit (GPU), field programmable gate array (FPGA), application specific integrated circuit (ASIC), or other programmable hardware device that allow programming of network interface 2400. For example, a “smart network interface” or SmartNIC can provide packet processing capabilities in the network interface using processors 2404. In some examples, processors 2404 can be implemented as a processor component for a SmartNIC.

[00139] Packet allocator 2424 can provide distribution of received packets for processing by multiple CPUs or cores using timeslot allocation described herein or RSS. When packet allocator 2424 uses RSS, packet allocator 2424 can calculate a hash or make another determination based on contents of a received packet to determine which CPU or core is to process a packet.

[00140] Interrupt coalesce 2422 can perform interrupt moderation whereby network interface interrupt coalesce 2422 waits for multiple packets to arrive, or for a time-out to expire, before generating an interrupt to host system to process received packet(s). Receive Segment Coalescing (RSC) can be performed by network interface 2400 whereby portions of incoming packets are combined into segments of a packet. Network interface 2400 provides this coalesced packet to an application.

[00141] Direct memory access (DMA) engine 2452 can copy a packet header, packet payload, and/or descriptor directly from host memory to the network interface or vice versa, instead of copying the packet to an intermediate buffer at the host and then using another copy operation from the intermediate buffer to the destination buffer. In some embodiments, multiple DMA engines are available for transfer of contents of packets to a destination memory associated with a host device or a destination memory associated with an accelerator device.

[00142] Memory 2410 can be any type of volatile or non-volatile memory device and can store any queue or instructions used to program network interface 2400. Transmit queue 2406 can include data or references to data for transmission by network interface. Receive queue 2408 can include data or references to data that was received by network interface from a network. Descriptor queues 2420 can include descriptors that reference data or packets in transmit queue 2406 or receive queue 2408 and corresponding destination memory regions. Bus interface 2412 can provide an interface with host device (not depicted). For example, bus interface 2412 can be compatible with PCI, PCI Express, PCI-x, Serial ATA, and/or USB compatible interface (although other interconnection standards may be used).

[00143] FIG. 25 depicts a system. The system can use embodiments described herein in connection with requesting a DMA operation, performing a DMA operation, accessing a remote memory device, or prefetching content as described herein. System 2500 includes processor 2510, which provides processing, operation management, and execution of instructions for system 2500. Processor 2510 can include any type of microprocessor, central processing unit (CPU), graphics processing unit (GPU), processing core, or other processing hardware to provide processing for system 2500, or a combination of processors. Processor 2510 controls the overall operation of system 2500, and can be or include, one or more programmable general-purpose or special-purpose microprocessors, digital signal processors (DSPs), programmable controllers, application specific integrated circuits (ASICs), programmable logic devices (PUDs), or the like, or a combination of such devices.

[00144] In one example, system 2500 includes interface 2512 coupled to processor 2510, which can represent a higher speed interface or a high throughput interface for system components that needs higher bandwidth connections, such as memory subsystem 2520, graphics interface components 2540, or accelerators 2542. Interface 2512 represents an interface circuit, which can be a standalone component or integrated onto a processor die. Where present, graphics interface 2540 interfaces to graphics components for providing a visual display to a user of system 2500. In one example, graphics interface 2540 can drive a high definition (HD) display that provides an output to a user. High definition can refer to a display having a pixel density of approximately 100 PPI (pixels per inch) or greater and can include formats such as full HD (e.g., 1080p), retina displays, 4K (ultra-high definition or UHD), or others. In one example, the display can include a touchscreen display. In one example, graphics interface 2540 generates a display based on data stored in memory 2530 or based on operations executed by processor 2510 or both. In one example, graphics interface 2540 generates a display based on data stored in memory 2530 or based on operations executed by processor 2510 or both.

[00145] Accelerators 2542 can be a programmable or fixed function offload engine that can be accessed or used by a processor 2510. For example, an accelerator among accelerators 2542 can provide compression (DC) capability, cryptography services such as public key encryption (PKE), cipher, hash/authentication capabilities, decryption, or other capabilities or services. In some embodiments, in addition or alternatively, an accelerator among accelerators 2542 provides field select controller capabilities as described herein. In some cases, accelerators 2542 can be integrated into a CPU socket (e.g., a connector to a motherboard or circuit board that includes a CPU and provides an electrical interface with the CPU). For example, accelerators 2542 can include a single or multi-core processor, graphics processing unit, logical execution unit single or multi-level cache, functional units usable to independently execute programs or threads, application specific integrated circuits (ASICs), neural network processors (NNPs), programmable control logic, and programmable processing elements such as field programmable gate arrays (FPGAs). Accelerators 2542 can provide multiple neural networks, processor cores, or graphics processing units can be made available for use by artificial intelligence (Al) or machine learning (MU) models. For example, the Al model can use or include any or a combination of: a reinforcement learning scheme, Q-leaming scheme, deep-Q learning, or Asynchronous Advantage Actor-Critic (A3C), combinatorial neural network, recurrent combinatorial neural network, or other Al or MU model. Multiple neural networks, processor cores, or graphics processing units can be made available for use by Al or MU models.

[00146] Memory subsystem 2520 represents the main memory of system 2500 and provides storage for code to be executed by processor 2510, or data values to be used in executing a routine. Memory subsystem 2520 can include one or more memory devices 2530 such as read-only memory (ROM), flash memory, one or more varieties of random access memory (RAM) such as DRAM, or other memory devices, or a combination of such devices. Memory 2530 stores and hosts, among other things, operating system (OS) 2532 to provide a software platform for execution of instructions in system 2500. Additionally, applications 2534 can execute on the software platform of OS 2532 from memory 2530. Applications 2534 represent programs that have their own operational logic to perform execution of one or more functions. Processes 2536 represent agents or routines that provide auxiliary functions to OS 2532 or one or more applications 2534 or a combination. OS 2532, applications 2534, and processes 2536 provide software logic to provide functions for system 2500. In one example, memory subsystem 2520 includes memory controller 2522, which is a memory controller to generate and issue commands to memory 2530. It will be understood that memory controller 2522 could be a physical part of processor 2510 or a physical part of interface 2512. For example, memory controller 2522 can be an integrated memory controller, integrated onto a circuit with processor 2510.

[00147] In some examples, OS 2532 can determine a capability of a device associated with a device driver. For example, OS 2532 can receive an indication of a capability of a device (e.g., NIC 2550 or a storage configuration interface) to configure a NIC 2550 to perform any of the capabilities described herein (e.g., NIC 2550 retrieving memory access commands from a region of memory or performing prefetching). OS 2532 can request a driver to enable or disable NIC 2550 to perform any of the capabilities described herein. In some examples, OS 2532, itself, can enable or disable NIC 2550 to perform any of the capabilities described herein. OS 2532 can provide requests (e.g., from an application or VM) to NIC 2550 to utilize one or more capabilities of NIC 2550. For example, any application can request use or non-use of any of capabilities described herein by NIC 2550.

[00148] While not specifically illustrated, it will be understood that system 2500 can include one or more buses or bus systems between devices, such as a memory bus, a graphics bus, interface buses, or others. Buses or other signal lines can communicatively or electrically couple components together, or both communicatively and electrically couple the components. Buses can include physical communication lines, point-to-point connections, bridges, adapters, controllers, or other circuitry or a combination. Buses can include, for example, one or more of a system bus, a Peripheral Component Interconnect (PCI) bus, a HyperTransport or industry standard architecture (ISA) bus, a small computer system interface (SCSI) bus, a universal serial bus (USB), or an Institute of Electrical and Electronics Engineers (IEEE) standard 1394 bus.

[00149] In one example, system 2500 includes interface 2514, which can be coupled to interface 2512. In one example, interface 2514 represents an interface circuit, which can include standalone components and integrated circuitry. In one example, multiple user interface components or peripheral components, or both, couple to interface 2514. Network interface 2550 provides system 2500 the ability to communicate with remote devices (e.g., servers or other computing devices) over one or more networks. Network interface 2550 can include an Ethernet adapter, wireless interconnection components, cellular network interconnection components, USB (universal serial bus), or other wired or wireless standards-based or proprietary interfaces. Network interface 2550 can transmit data to a remote device, which can include sending data stored in memory. Network interface 2550 can receive data from a remote device, which can include storing received data into memory. Various embodiments can be used in connection with network interface 2550, processor 2510, and memory subsystem 2520.

[00150] In one example, system 2500 includes one or more input/output (I/O) interface(s) 2560. I/O interface 2560 can include one or more interface components through which a user interacts with system 2500 (e.g., audio, alphanumeric, tactile/touch, or other interfacing). Peripheral interface 2570 can include any hardware interface not specifically mentioned above. Peripherals refer generally to devices that connect dependently to system 2500. A dependent connection is one where system 2500 provides the software platform or hardware platform or both on which operation executes, and with which a user interacts.

[00151] In one example, system 2500 includes storage subsystem 2580 to store data in a nonvolatile manner. In one example, in certain system implementations, at least certain components of storage 2580 can overlap with components of memory subsystem 2520. Storage subsystem 2580 includes storage device(s) 2584, which can be or include any conventional medium for storing large amounts of data in a nonvolatile manner, such as one or more magnetic, solid state, or optical based disks, or a combination. Storage 2584 holds code or instructions and data 2586 in a persistent state (e.g., the value is retained despite interruption of power to system 2500). Storage 2584 can be generically considered to be a "memory," although memory 2530 is typically the executing or operating memory to provide instructions to processor 2510. Whereas storage 2584 is nonvolatile, memory 2530 can include volatile memory (e.g., the value or state of the data is indeterminate if power is interrupted to system 2500). In one example, storage subsystem 2580 includes controller 2582 to interface with storage 2584. In one example controller 2582 is a physical part of interface 2514 or processor 2510 or can include circuits or logic in both processor 2510 and interface 2514.

[00152] A volatile memory is memory whose state (and therefore the data stored in it) is indeterminate if power is interrupted to the device. Dynamic volatile memory requires refreshing the data stored in the device to maintain state. One example of dynamic volatile memory incudes DRAM (Dynamic Random Access Memory), or some variant such as Synchronous DRAM (SDRAM). Another example of volatile memory includes cache or static random access memory (SRAM). A memory subsystem as described herein may be compatible with a number of memory technologies, such as DDR3 (Double Data Rate version 3, original release by JEDEC (Joint Electronic Device Engineering Council) on June 27, 2007). DDR4 (DDR version 4, initial specification published in September 2012 by JEDEC), DDR4E (DDR version 4), LPDDR3 (Low Power DDR version3, JESD209-3B, August 2013 by JEDEC), LPDDR4) LPDDR version 4, JESD209-4, originally published by JEDEC in August 2014), WIO2 (Wide Input/output version 2, JESD229-2 originally published by JEDEC in August 2014, HBM (High Bandwidth Memory, JESD325, originally published by JEDEC in October 2013, LPDDR5 (currently in discussion by JEDEC), HBM2 (HBM version 2), currently in discussion by JEDEC, or others or combinations of memory technologies, and technologies based on derivatives or extensions of such specifications. The JEDEC standards are available at www.jedec.org.

[00153] A non-volatile memory (NVM) device is a memory whose state is determinate even if power is interrupted to the device. In one embodiment, the NVM device can comprise a block addressable memory device, such as NAND technologies, or more specifically, multi-threshold level NAND flash memory (for example, Single-Level Cell (“SLC”), Multi-Level Cell (“MLC”), Quad-Level Cell (“QLC”), Tri-Level Cell (“TLC”), or some other NAND). A NVM device can also comprise a byte-addressable write-in-place three dimensional cross point memory device, or other byte addressable write-in-place NVM device (also referred to as persistent memory), such as single or multi-level Phase Change Memory (PCM) or phase change memory with a switch (PCMS), Intel® Optane™ memory, NVM devices that use chalcogenide phase change material (for example, chalcogenide glass), resistive memory including metal oxide base, oxygen vacancy base and Conductive Bridge Random Access Memory (CB-RAM), nanowire memory, ferroelectric random access memory (FeRAM, FRAM), magneto resistive random access memory (MRAM) that incorporates memristor technology, spin transfer torque (STT)-MRAM, a spintronic magnetic junction memory based device, a magnetic tunneling junction (MTJ) based device, a DW (Domain Wall) and SOT (Spin Orbit Transfer) based device, a thyristor based memory device, or a combination of any of the above, or other memory.

[00154] A power source (not depicted) provides power to the components of system 2500. More specifically, power source typically interfaces to one or multiple power supplies in system 2500 to provide power to the components of system 2500. In one example, the power supply includes an AC to DC (alternating current to direct current) adapter to plug into a wall outlet. Such AC power can be renewable energy (e.g., solar power) power source. In one example, power source includes a DC power source, such as an external AC to DC converter. In one example, power source or power supply includes wireless charging hardware to charge via proximity to a charging field. In one example, power source can include an internal battery, alternating current supply, motion-based power supply, solar power supply, or fuel cell source.

[00155] In an example, system 2500 can be implemented using interconnected compute sleds of processors, memories, storages, network interfaces, and other components. High speed connections can be used such as: Ethernet (IEEE 802.3), remote direct memory access (RDMA), InfiniBand, Internet Wide Area RDMA Protocol (iWARP), Transmission Control Protocol (TCP), User Datagram Protocol (UDP), quick UDP Internet Connections (QUIC), RDMA over Converged Ethernet (RoCE), Peripheral Component Interconnect express (PCIe), Intel® QuickPath Interconnect (QPI), Intel® Ultra Path Interconnect (UPI), Intel® On-Chip System Fabric (IOSF), Omnipath, Compute Express Link (CXL), HyperTransport, highspeed fabric, NVLink, Advanced Microcontroller Bus Architecture (AMBA) interconnect, OpenCAPI, Gen-Z, Cache Coherent Interconnect for Accelerators (CCIX), 3GPP Long Term Evolution (LTE) (4G), 3GPP 5G, DisplayPort, embedded DisplayPort, MIPI, HDMI, Infinity Fabric (IF), and successors or variations thereof.

[00156] Embodiments herein may be implemented in various types of computing and networking equipment, such as switches, routers, racks, and blade servers such as those employed in a data center and/or server farm environment. The servers used in data centers and server farms comprise arrayed server configurations such as rack-based servers or blade servers. These servers are interconnected in communication via various network provisions, such as partitioning sets of servers into Local Area Networks (LANs) with appropriate switching and routing facilities between the LANs to form a private Intranet. For example, cloud hosting facilities may typically employ large data centers with a multitude of servers. A blade comprises a separate computing platform that is configured to perform server-type functions, that is, a “server on a card.” Accordingly, each blade includes components common to conventional servers, including a main printed circuit board (main board) providing internal wiring (e.g., buses) for coupling appropriate integrated circuits (ICs) and other components mounted to the board.

[00157] In some examples, network interface and other embodiments described herein can be used in connection with a base station (e.g., 3G, 4G, 5G and so forth), macro base station (e.g., 5G networks), picostation (e.g., an IEEE 802.11 compatible access point), nanostation (e.g., for Point-to-MultiPoint (PtMP) applications), on-premises data centers, off-premises data centers, edge network elements, fog network elements, and/or hybrid data centers (e.g., data center that use virtualization, cloud and software- defined networking to deliver application workloads across physical data centers and distributed multicloud environments).

[00158] Various examples may be implemented using hardware elements, software elements, or a combination of both. In some examples, hardware elements may include devices, components, processors, microprocessors, circuits, circuit elements (e.g., transistors, resistors, capacitors, inductors, and so forth), integrated circuits, ASICs, PLDs, DSPs, FPGAs, memory units, logic gates, registers, semiconductor device, chips, microchips, chip sets, and so forth. In some examples, software elements may include software components, programs, applications, computer programs, application programs, system programs, machine programs, operating system software, middleware, firmware, software modules, routines, subroutines, functions, methods, procedures, software interfaces, APIs, instruction sets, computing code, computer code, code segments, computer code segments, words, values, symbols, or any combination thereof. Determining whether an example is implemented using hardware elements and/or software elements may vary in accordance with any number of factors, such as desired computational rate, power levels, heat tolerances, processing cycle budget, input data rates, output data rates, memory resources, data bus speeds and other design or performance constraints, as desired for a given implementation. A processor can be one or more combination of a hardware state machine, digital control logic, central processing unit, or any hardware, firmware and/or software elements.

[00159] Some examples may be implemented using or as an article of manufacture or at least one computer-readable medium. A computer-readable medium may include a non-transitory storage medium to store logic. In some examples, the non-transitory storage medium may include one or more types of computer-readable storage media capable of storing electronic data, including volatile memory or nonvolatile memory, removable or non-removable memory, erasable or non-erasable memory, writeable or rewriteable memory, and so forth. In some examples, the logic may include various software elements, such as software components, programs, applications, computer programs, application programs, system programs, machine programs, operating system software, middleware, firmware, software modules, routines, subroutines, functions, methods, procedures, software interfaces, API, instruction sets, computing code, computer code, code segments, computer code segments, words, values, symbols, or any combination thereof.

[00160] According to some examples, a computer-readable medium may include a non-transitory storage medium to store or maintain instructions that when executed by a machine, computing device or system, cause the machine, computing device or system to perform methods and/or operations in accordance with the described examples. The instructions may include any suitable type of code, such as source code, compiled code, interpreted code, executable code, static code, dynamic code, and the like. The instructions may be implemented according to a predefined computer language, manner or syntax, for instructing a machine, computing device or system to perform a certain function. The instructions may be implemented using any suitable high-level, low-level, object-oriented, visual, compiled and/or interpreted programming language.

[00161] One or more aspects of at least one example may be implemented by representative instructions stored on at least one machine-readable medium which represents various logic within the processor, which when read by a machine, computing device or system causes the machine, computing device or system to fabricate logic to perform the techniques described herein. Such representations, known as “IP cores” may be stored on a tangible, machine readable medium and supplied to various customers or manufacturing facilities to load into the fabrication machines that actually make the logic or processor.

[00162] The appearances of the phrase “one example” or “an example” are not necessarily all referring to the same example or embodiment. Any aspect described herein can be combined with any other aspect or similar aspect described herein, regardless of whether the aspects are described with respect to the same figure or element. Division, omission or inclusion of block functions depicted in the accompanying figures does not infer that the hardware components, circuits, software and/or elements for implementing these functions would necessarily be divided, omitted, or included in embodiments.

[00163] Some examples may be described using the expression "coupled" and "connected" along with their derivatives. These terms are not necessarily intended as synonyms for each other. For example, descriptions using the terms “connected” and/or “coupled” may indicate that two or more elements are in direct physical or electrical contact with each other. The term "coupled,” however, may also mean that two or more elements are not in direct contact with each other, but yet still co-operate or interact with each other.

[00164] The terms “first,” “second,” and the like, herein do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The terms “a” and “an” herein do not denote a limitation of quantity, but rather denote the presence of at least one of the referenced items. The term “asserted” used herein with reference to a signal denote a state of the signal, in which the signal is active, and which can be achieved by applying any logic level either logic 0 or logic 1 to the signal. The terms “follow” or “after” can refer to immediately following or following after some other event or events. Other sequences of steps may also be performed according to alternative embodiments. Furthermore, additional steps may be added or removed depending on the particular applications. Any combination of changes can be used and one of ordinary skill in the art with the benefit of this disclosure would understand the many variations, modifications, and alternative embodiments thereof.

[00165] Disjunctive language such as the phrase “at least one of X, Y, or Z,” unless specifically stated otherwise, is otherwise understood within the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (e.g., X, Y, and/or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one ofY, or at least one of Z to each be present. Additionally, conjunctive language such as the phrase “at least one of X, Y, and Z,” unless specifically stated otherwise, should also be understood to mean X, Y, Z, or any combination thereof, including “X, Y, and/or Z

[00166] An example includes pre-fetching content from one or more memory devices based on an anticipated next executed function in a sequence of functions. In some examples, content includes data or executable code. In some examples, content is pre-fetched prior to a request to retrieve the content is made. In some examples, a network interface can perform the pre-fetching.

[00167] Example 1 includes a method comprising: at a direct memory access (DMA) device: accessing a command from a memory region allocated to receive commands for execution by the DMA device, wherein the command is to access content from a local memory device or remote memory node.

[00168] Example 2 includes any example, wherein the content comprises one or more of: data and executable instructions.

[00169] Example 3 includes any example, and includes determining if the content is stored in a local memory device or a remote memory node based on a configuration that indicates whether a source address associated with the command refers to a memory address associated with the local memory device or the remote memory node and whether a destination address refers to a memory address associated with the local memory device or the remote memory node; based on the content being stored in a local memory device, copying the content from the local memory device to the DMA device for transmission to a remote memory node; and based on the content being stored in a remote memory node: causing transmission of a request for content to the remote memory node and copying received content from the remote memory node to the local memory device.

[00170] Example 4 includes any example, and includes the DMA device copying the content from a local memory device or copying the content to the local memory device using a memory interface.

[00171] Example 5 includes any example, wherein the memory interface comprises one or more of Compute Express Link or Double Data Rate (DDRx) where x is an integer.

[00172] Example 6 includes any example, and includes at the DMA device: receiving an identification of one or more functions in a function sequence; receiving configuration of content to access for utilization in each of the one or more functions in the sequence; and based on an indication that a particular function is to execute, pre-fetching content associated with the function. [00173] Example 7 includes any example, wherein the content associated with the function comprises one or more of execution code or data.

[00174] Example 8 includes any example, and includes a computer-readable medium comprising instructions stored thereon, that if executed by one or more processors, cause the one or more processors to: request a direct memory access (DMA) operation by writing the DMA operation to a memory region allocated to receive commands for execution by a DMA device, wherein the commands are associated with content access operations.

[00175] Example 9 includes any example, wherein: based on the content being stored in a local memory device, the DMA operation is to cause copying the content from the local memory device to the DMA device for transmission to a remote memory node.

[00176] Example 10 includes any example, wherein: based on the content being stored in a remote memory node, the DMA operation is to cause transmission of a request for content to the remote memory node and copying received content from the remote memory node to a local memory device.

[00177] Example 11 includes any example, wherein content is copied to the local memory device from the DMA device using a memory interface.

[00178] Example 12 includes any example, wherein the content comprises one or more of execution code or data.

[00179] Example 13 includes any example, and includes an apparatus comprising: at least one processor and a direct memory access (DMA) device, wherein the DMA device is to: access a command from a memory region allocated to receive commands for execution by the DMA device, wherein the command is to access content from a local memory device or remote memory node.

[00180] Example 14 includes any example, wherein the DMA device is to: determine if the content is stored in a local memory device or a remote memory node based on a configuration that indicates whether a source address refers to a memory address associated with the local memory device or the remote memory node and whether a destination address refers to a memory address associated with the local memory device or the remote memory node.

[00181] Example 15 includes any example, wherein the DMA device is to: copy the content from a local memory device or copy the content to the local memory device using a memory interface.

[00182] Example 16 includes any example, wherein the memory interface comprises one or more of Compute Express Link or Double Data Rate (DDRx) where x is an integer.

[00183] Example 17 includes any example, wherein the DMA device is to: receive an identification of one or more functions in a sequence; receive configuration of content to access for utilization in each of the one or more functions in the sequence; and based on an indication that a particular function is to execute, retrieve content associated with the function.

[00184] Example 18 includes any example, wherein the content associated with the function comprises one or more of execution code or data. [00185] Example 19 includes any example, and includes a local memory coupled to the DMA device using a memory interface.

[00186] Example 20 includes any example, wherein the at least one processor is part of a server, rack of servers, or data center and one or more of the server, rack of servers, or data center is to access the content from the local memory device or remote memory node.