BACKGROUND

[0001] Data centers and other large-scale computing facilities can include distributed computing systems having thousands and even millions of servers interconnected by network routers, switches, bridges, load balancers, firewalls, and other types of network devices. Such servers, network devices, and other types of electronic components are typically housed in large buildings, rooms, racks, cabinets, containers, or other suitable enclosures. The individual servers can host virtual machines, virtual switches, or other types of virtualized components cooperating with one another to execute applications in order to provide computing services to users.

SUMMARY

[0002] This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

[0003] For proper operation of servers and other electronic components, computing facilities typically include an environmental conditioning system to maintain buildings or other types of enclosures housing the electronic components at suitable temperature and/or humidity levels. For example, a heating, ventilation, and air conditioning (“HVAC”) system can be configured to provide cooling air to enclosures to remove heat generated by electronic components housed in the enclosures. HVAC systems can typically include cooling towers, air or water cooled chillers, fans, blowers, heat exchangers, air dampers, louvers, and/or other suitable components. For instance, an HVAC system can include an air handling unit having one or more air movers (e.g., fans) and a heat exchanger in fluid communication with a source of cooling fluid (e.g., cooling water, chilled water, etc.). During operation, cooling air can enter an enclosure from an air inlet, past operating electronic components in the enclosure to remove heat therefrom, and exhaust as hot air via an air outlet into a return air plenum. The fan in the air handling unit can then force the hot air carrying the absorbed heat from the electronic components to pass through the heat exchanger and dissipate the absorbed heat to the cooling fluid. The cooled air can then be circulated back as cooling air to the enclosure to continually cool the electronic components. As such, by adjusting an amount and/or temperature of cooling air, cooling fluid, etc., the HVAC system can provide an operating environment in the enclosure to allow proper operations of the electronic components housed in the enclosure.

[0004] The foregoing heat removal technique in computing facilities can have several drawbacks. First, during normal operation, an HVAC system can adequately remove heat from the electronic components housed in the enclosure. However, when the HVAC system experiences a failure due to, for instance, an air handling unit failure, a chiller failure, a control valve malfunction, a plugged strainer, or other types of failures. Such cooling interruption can cause temperatures of the electronic components to rise to unsafe levels within a short period, for example, even five to ten minutes for switching the HVAC system to a backup diesel generator. Secondly, the HVAC system is typically sized to have a cooling capacity corresponding to a peak load in the computing facility to ensure full availability of computing capacity. However, the computing facility may only be operating at the peak load infrequently. As such, the cooling capacity in the HVAC system may be underutilized most of the time. Such underutilization of the HVAC system can render high costs of providing such a cooling capacity wasteful.

[0005] Several embodiments of the disclosed technology can address at least certain aspects of the foregoing drawbacks by implementing a phase change material (“PCM”) into buildings, rooms, racks, cabinets, containers, or other suitable types of enclosures housing servers or other suitable types of electronic components. As used herein, a phase change material generally refers to a substance having a composition corresponding to a designed melting point. The melting point of the phase change material may be adjusted by varying a composition of the substance via, for example, hydrogenation, esterification, and/or mixing with additives. For example, pure coconut oil which typically has a melting point of about 24°C may be hydrogenated to have a higher melting point of 25°C, 3 CPC, 35°C, 45°C, 50°C, or 55°C. In another example, glycol, ethylene glycol, propylene glycol, and/or other suitable anti-freeze agents may be used as additives to, for instance, a hydrogenated coconut oil to reduce the melting point to about 25°C, 30°C, 35°C, or 50°C. In further examples, the phase change material can also include pure or hydrogenated palm oil, cottonseed oil, and/or other suitable types of oils, lipids, or other suitable materials.

[0006] The phase change material may be incorporated into a computing facility to provide standby cooling at or above the melting point of the phase change material. For example, in certain embodiments, a phase change material can be incorporated into an air inlet of an enclosure such as a rack or cabinet housing multiple servers. The phase change material can have a suitable composition configured to have a melting point of, for instance, about 25°C to about 3 CPC, which can be equal to or above normal operating temperatures inside the enclosure. During normal operation, the HVAC system maintains a temperature of air entering an air inlet of the enclosure housing the electronic components at below the melting point of the phase change material. As such, the phase change material at the air inlet would stay solid.

[0007] During a cooling outage of the HVAC system, the temperature of air entering the enclosure can rise to above the melting point of the phase change material, e.g., about 25°C to about 30°C. At the elevated temperature, the phase change material at the air inlet starts to transition from a solid to a two-phase substance, and eventually to a liquid while absorbing heat (e.g., specific and latent heat) from the air entering the enclosure, thereby reducing the temperature of the incoming air before the air contacts the electronic components. As such, the phase change material can provide an amount of standby cooling to the electronic components in the enclosure during outage of the HVAC system. Subsequently, the HVAC system may be restarted by, for instance, correcting an air handling unit failure, a chiller failure, a control valve malfunction, a strainer being plugged, or other failures. As a result, cooling air from the HVAC system can reduce the temperature of the air entering the enclosure to below 30°C. The phase change material can eject heat to the now lower-temperature cooling air and reverse back from a liquid to a solid, thereby restoring the standby cooling capacity.

[0008] Various techniques may be practiced to incorporate the phase change material into the enclosure housing the electronic components. In one example, a phase-change heat exchanger can be incorporated into the air inlet of the enclosure. Each of the tubes in the heat exchanger can be constructed with a heat conducting material (e.g., copper or aluminum) and can optionally include fins, baffles, or other suitable heat enhancing components. Each of the tubes can also contain a select amount of the phase change material. During normal operation, cooling air at a temperature lower than the melting point of the phase change material can pass through the tubes without melting the phase change material inside the tubes. During an outage of the HVAC system, air entering the enclosure may be at an elevated temperature equal to or higher than the melting point of the phase change material in the tubes. As such, heat can be transferred from the incoming air to the phase change material in the tubes and resulting in melting the phase change material while lowering the temperature of the incoming air by, for instance, 5°C to 10°C. As such, air entering the enclosure can be maintained at a cooling temperature (e.g., about 25°C, to about 30°C) for a period of time until the HVAC system is restarted by, for instance, correcting an air handling unit failure, a chiller failure, a control valve malfunction, a strainer being plugged, or other failures.

[0009] In other embodiments, the phase change material can also be incorporated into a room, a rack, a container, or a building in the computing facility in addition to or in lieu of being incorporated into individual enclosures housing the electronic components. For instance, the phase change material can be incorporated into celling panels, floor boards, or other suitable locations in a room or building in the computing facility. During normal operation, the phase change material can be solid when the HVAC system maintains the room temperature below the melting point of the phase change material. During an outage of the HVAC system, the room temperature would rise. When the room temperature rises to be equal to or above the melting point of the phase change material, the phase change material would start melting by absorbing latent heat from air in the room, rack, container, or building, and thus provide standby cooling to the computing facility.

[0010] Several embodiments of the disclosed technology can efficiently accommodate cooling outages in the HVAC system by providing standby cooling to the electronic devices using a phase change material. During a cooling outage of the HVAC system, the phase change material can absorb heat from air entering the enclosure once a temperature of the incoming air is above a melting point of the phase change material. As such, the phase change material can provide uninterruptable cooling to the electronic components housed in the enclosure. Also, the phase change material can provide an amount of stored cooling capacity in the computing facility such that the HVAC system may be sized to have a capacity corresponding to a normal load of the computing facility instead of a peak load. During a peak load, the phase change material can provide an amount of cooling capacity to supplement that from the HVAC system. As such, capital investment and operating costs of the HVAC system may be reduced.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1A is a schematic diagram of a computing facility having standby cooling supply during normal operation in accordance with embodiments of the disclosed technology.

[0012] Figure IB is a schematic diagram of the computing facility of Figure 1 A during a cooling outage in accordance with embodiments of the disclosed technology.

(0013] Figure 2A is a schematic front cross-sectional view of a heat exchanger having a phase change material incorporated into an enclosure housing multiple computing units in accordance with additional embodiments of the disclosed technology.

[0014] Figure 2B is a schematic top cross-sectional view of the enclosure in Figure 2A in accordance with additional embodiments of the disclosed technology.

[0015] Figures 3A-3C are schematic diagrams illustrating certain components and configurations of a source of cooling fluid suitable for the computing facility in Figure 1 A in accordance with embodiments of the disclosed technology.

DETAILED DESCRIPTION

[0016] Certain embodiments of systems, devices, components, modules, routines, and processes for providing standby cooling to electronic components housed in enclosures of a computing facility are described below. In the following description, specific details of components are included to provide a thorough understanding of certain embodiments of the disclosed technology. A person skilled in the relevant art can also understand that the disclosed technology may have additional embodiments or may be practiced without several of the details of the embodiments described below with reference to Figures 1 A-3C.

[0017] As used herein, a “computing facility” generally refers to a computing system having a plurality of network devices that interconnect a plurality of servers to one another or to external networks (e.g., the Internet) as well as buildings, rooms, containers, racks, cabinets, or other suitable types of enclosures that house the servers, the network devices, or other suitable types of electronic components. The term “network device” generally refers to a physical network device, examples of which include routers, switches, hubs, bridges, load balancers, security gateways, or firewalls. A “computing unit” generally refers to a server or other suitable types of computing device configured to implement, for instance, one or more virtual machines or other suitable virtualized components.

[0018] Also used herein, a “heating, ventilation, and air conditioning” or “HVAC” system generally refers to a system having components configured to provide a suitable operating environment to electronic components housed in enclosures in a computing facility. In one example, an HVAC system can include one or more air movers configured to force air through a heat exchanger circulated with a cooling fluid to produce cooling air for entering an enclosure housing servers. In another example, an HVAC system can also include a chiller or other suitable refrigeration equipment configured to provide chilled air or chilled water to the enclosure housing the electronic components. In further examples, an HVAC system can also include cooling towers, moisture absorbers, and/or other suitable equipment. Specific examples of HVAC system are described in more detail below with reference to Figures 1A-3C.

[0019] The term “air mover” generally refers to an apparatus configured to generate an air flow. Examples of air movers can include fans, blowers, or compressors having multiple blades configured to push, pull, or otherwise impart movement to air in order to generate an air flow. An air mover can include a single stage or multi-stage fan, blower, or compressor. One example fan suitable for generating an air flow is an adjustable propeller fan provided by Aerovent of Minneapolis, Minnesota. Other examples of air movers can include centrifugal fans/blowers, tube-axial fans, vane-axial fans, or other suitable devices. [0020] A “phase change material” or “PCM” generally refers to a substance having a composition corresponding to a designed melting point such that the substance is a solid during one operating mode and a liquid during a different operating mode of the electronic components in the computing facility. The melting point of the phase change material may be adjusted by varying a composition of the substance via, for example, hydrogenation and/or adding additives. For example, pure coconut oil which typically has a melting point of about 24°C may be hydrogenated to have a higher melting point of 25°C, 3 CPC, 35°C, 40°C, 45°C, 50°C, or 55°C. In another example, glycol, ethylene glycol, propylene glycol, and/or other suitable anti-freeze agents may be added to, for instance, a hydrogenated coconut oil to reduce the melting point to about 25°C to about 3 CPC. In further examples, the phase change material can also include hydrogenated palm oil, cottonseed oil, and/or other suitable types of oils, lipids, or other suitable materials.

[0021] For proper operation of servers and other electronic components, computing facilities typically include an HVAC system to maintain buildings or other types of enclosures housing the electronic components at suitable temperature and/or humidity levels. For instance, an example HVAC system can include an air handling unit having one or more air movers (e.g., fans) and a heat exchanger in fluid communication with a source of a cooling fluid (e.g., cooling water, chilled water, etc.). During operation, cool air can enter an enclosure from an air inlet, past operating electronic components in the enclosure to remove heat therefrom, and exhaust as hot air via an air outlet into a return air plenum. The fan in the air handling unit can then force the hot air carrying the absorbed heat from the electronic components to pass through the heat exchanger and dissipate the absorbed heat to the cooling fluid. The cooled air can then be circulated back to the enclosure to continually cooling the electronic components. As such, by adjusting an amount and/or temperature of cooling air, cooling fluid, etc., the HVAC system can provide an operating environment in the enclosure to allow proper operations of the electronic components housed in the enclosure.

[0022] The foregoing heat removal technique using an HVAC system in computing facilities can have several drawbacks. First, during normal operation, the HVAC system can adequately remove heat from the electronic components housed in the enclosure. However, when the HVAC system experiences a cooling outage due to, for instance, a chiller failure, a control valve malfunction, a plugged strainer, or other types of failures, cooling air provided to the enclosure may be interrupted. Such interruption in the provided cooling can cause temperatures of the electronic components to rise to unsafe levels within a short period of time, especially during peak loading. Secondly, the HVAC system is typically sized to have a cooling capacity corresponding to a peak load in the computing facility to ensure full availability of computing capacity. However, the computing facility may only be operating at the peak load infrequently. As such, the cooling capacity in the HVAC system may be underutilized most of the time. Such underutilization of the HVAC system can render high costs of providing such a cooling capacity wasteful.

[0023j Several embodiments of the disclosed technology can address at least certain aspects of the foregoing drawbacks by implementing a phase change material into buildings, rooms, racks, cabinets, or other suitable types of enclosures housing the electronic components. In one example, a phase-change heat exchanger can be incorporated into the air inlet of the enclosure. Each of the tubes in the heat exchanger can contain a select amount of the phase change material. During normal operation, cooling air at a temperature lower than the melting point (e.g., about 25°C to about 30°C) of the phase change material can pass through the tubes without melting the phase change material. During a cooling outage, air entering the enclosure may be at an elevated temperature equal to or higher than the melting point of the phase change material in the tubes. As such, heat can be transferred from the incoming air to the phase change material in the tubes to resulting in melting of the phase change material and a reduction of the temperature of the incoming air by, for instance, 5°C to 10°C. As such, air entering the enclosure can be maintained at a reduced temperature (e.g., 25°C) for a period of time until the HVAC system is restarted by, for instance, correcting a chiller failure, a control valve malfunction, a plugged strainer, or other types of failures, as described below with reference to Figures 1 A-3C.

|0024| Figure 1A is a schematic diagram of a computing facility 100 having standby cooling supply during normal operation in accordance with embodiments of the disclosed technology. As shown in Figure 1 A, the computing facility 100 can include a structure 102 (e.g., a building, a container, etc.) with one or more enclosures 104 individually holding multiple computing units 101 and one or more air handling units 106 (two are shown for illustration purposes). Two enclosures 104 are shown for illustration purposes though the computing facility 100 can be configured to accommodate any suitable numbers of enclosures 104 and/or computing units 101. Even though certain components of the computing facility 100 are shown in Figure 1A, in other embodiments, the computing facility 100 can also include other suitable electrical/mechanical components in similar or different arrangements.

[0025] As shown in Figure 1A, the computing facility 100 can include multiple computing units 104 coupled to one another by a computer network 108. The computer network 108 can include a wired medium (e.g., twisted pair, coaxial, untwisted pair, or optic fiber), a wireless medium (e.g., terrestrial microwave, cellular systems, WI-FI, wireless LANs, Bluetooth, infrared, near field communication, ultra-wide band, or free space optics), or a combination of wired and wireless media. The computer network 108 may operate according to Ethernet, token ring, asynchronous transfer mode, and/or other suitable protocols. In further embodiments, the computer network 108 can also include routers, switches, modems, and/or other suitable computing/communications components in any suitable arrangements.

[0026] The computing units 101 can be individually configured to implement one or more computing applications, network communications, input/output capabilities, and/or other suitable functionalities. In certain embodiments, the computing units 101 can include web servers, application servers, database servers, and/or other suitable computing components. In other embodiments, the processing units can include routers, network switches, analog/digital input/output modules, modems, and/or other suitable electronic components. Figure 1A shows four computing units 104 in each enclosure 104. In other embodiments, one, two, three, five, or any other suitable number of computing units 101 may be in each enclosure 104.

[0027] As shown in Figure 1A, the structure 102 can include multiple air spaces configured to circulate cooling air to the enclosures 104. For example, the structure 102 can include a cool air plenum 105a corresponding to each of the enclosure 104. The cool air plenum 105a is configured to receive cooling air 107 from a corresponding air handling unit 106 via, for instance, an input port 103a. The structure 102 can also include a hot air plenum 105b that is configured to receive exhaust air 107’ from the enclosures 104 after the cooling air 107 absorbs heat from the computing units 101 in order to provide cooling to the computing units 101. The structure 102 can further include a return air plenum 105c that is connected to the hot air plenum 105b via an outlet port 103b. The return air plenum 105c can be configured to route the exhaust air 107’ back to the air handling units 106, which in turn remove heat from the exhaust air 107’ to a cooling fluid (e.g., cooling water, chilled water, etc.) before recirculating the cooling air 107 to the cool air plenum 105a via the inlet port 103a.

[0028] The air handling units 106 can include suitable heat exchanging, air moving, moisture controlling, and/or other suitable types of devices. In the illustrated example, the air handling units 106 individually include a heat exchanger 111 and an air mover 110. The air mover 110 can be configured to force the exhaust air 107’ from the return air plenum 105c to be in thermal contact with the cooling fluid from a source 120 (shown in Figures 3 A-3C) of the cooling fluid. As such, the cooling fluid can absorb heat from the exhaust air 107’ to reduce a temperature of the exhaust air 107’ to, for instance, 25°C before being recirculated into the cool air plenum 105a as cooling air 107. In other examples, the air handling units 106 can also include multiple air movers arranged in an array, air louvers, air dampers, temperature sensors, and/or other suitable devices.

[0029] In normal operation, the air handling units 106 can provide sufficient cooling to the computing units 101 by ejecting heat absorbed by the cooling air 107 to the cooling fluid. However, under certain scenarios, the cooling capacities of the air handling units 106 may be diminished or completely lost. For example, the source 120 that provides the cooling fluid may experience a chiller failure, a control valve malfunction, a plugged strainer, or other types of failures. In other examples, a pump, a valve, or other suitable types of devices configured to circulate the cooling fluid may malfunction. In further examples, the individual air handling units 106 may fail. Under such scenarios, exhaust air 107’ at elevated temperatures (e.g., 25°C to 30°C) from the return air plenum 105c may be recirculated back to the enclosures 104 via the cool air plenum 105a. As such, the recirculated exhaust air 107’ may not provide sufficient cooling and consequently result in overheating of the computing units 101. As a result, performance of computing services provided by the computing units 101 may be degraded or even fail completely, and thus negatively impacting user experience.

[0030] Several embodiments of the disclosed technology can address certain aspects of the foregoing difficulty by providing standby cooling capacity to the enclosures 104 with a phase change material 116 (shown in Figure 2A). As shown in Figure 1A, each of the enclosures 104 can have an air inlet 104a in fluid communication with the cool air plenum 105a and an air outlet 104b in fluid communication with the hot air plenum 105b. At least proximate to the air inlet 104a, the enclosures 104 can each include a heat exchanger 112 having an air side in fluid communication with the cool air plenum 105a and a solid/liquid side containing a phase change material 116. [003JJUsing the phase change material 116, the heat exchangers 112 can be configured to provide standby cooling to the computing units 101 in the enclosures 104. In one embodiment, the phase change material 116 can be tuned via, for instance, varying a composition of the phase change material 116, to have a melting point that is equal to or above a normal operating temperature of the cooling air 107 entering the enclosures 104. For example, the cooling air 107 entering the enclosures 104 can have a normal operating temperature of 25°C while the phase change material 116 can have a melting point of about 25°C to about 30°C, about 25°C to about 35°C, about 25°C to about 40°C, or other suitable temperature ranges. As such, when the air handling units 106 provides sufficient cooling during normal operation, the cooling air 107 entering the enclosures 104 can pass through the heat exchangers 112 without melting the phase change material 116 in the solid/liquid side. As such, the phase change material 116 can stay as a solid (shown in Figure 1A as white lines with black background).

[0032} During a cooling outage, the phase change material 116 can provide an amount of standby cooling capacity to the computing units 101. For example, as shown in Figure IB, the cooling fluid to the heat exchangers 111 in the air handling units 106 may be interrupted (shown as dotted lines). As a result, the incoming air 107” entering the enclosures 104 via the cool air plenum 105a may be at an elevated temperature (e.g., 25°C to 30°C) that is equal to or above the melting point of the phase change material 116. When incoming air 107” contacts the heat exchanger 112, the phase change material 116 absorbs heat from the incoming air 117” to reduce the temperature of the incoming air 107” by a designed amount (e.g., 5°C to 10°C). and the absorbed heat can trigger a phase transition in the phase change material 116 from solid, to two-phase, and eventually to liquid. As such, the incoming air 107” may be at a lower temperature when contacting the computing units 101. The incoming air 107” at the lower temperature can remove heat from the computing units 101 and thus provide cooling to the computing units 101.

[0033} The amount of standby cooling capacity provided by the phase change material 116 can depend on a weight/volume of the phase change material 116 the heat exchangers 112 contain. The following is an example sizing calculation based on a 5KW power consumption by the computing units 101 and using coconut oil as the phase change material 116 for providing standby cooling for 30 minutes:

Q= 5 KW x 30 Minute = 2.5 KW Hr, (approximately 9000 KJ) where Q is total heat generated by the computing units 101. Based on the total heat, a mass (M) of the phase change material 116 can be calculated based on latent heat (L) of the phase change material 116 as:

M=Q/L= 9000 KJ / (110 KJ/Kg) = 81.8 Kg

Given a density (Ro) of coconut oil is 916 Kg/m ^A 3, the volume of the phase change material 116 can be calculated as:

V = M/Ro = 81.8 Kg / (916 Kg/m ^A 3) = 0.089 m ^A 3 Therefore, a total mass of the phase change material 116 is about 81.8 Kg with a volume of about 3.14 cubic feet. Though particular assumptions and values were used above for illustration purposes, in other implementations, similar calculations may be performed using other suitable assumptions and/or values for different power consumptions, phase change materials, etc.

[0034] Upon restoration of cooling to the air handling units 106, the cooling air 107 can again having a temperature that is below the melting point of the phase change material 116. As a result, the phase change material 116 can eject heat to the cooling air 107 and reverse the phase transition. As such, the phase change material 116 can refreeze and convert into a solid to regenerate the standby cooling capacity.

[0035] Various techniques may be practiced to incorporate the phase change material 116 into the heat exchangers 112. In one example, a phase-change heat exchanger 112 can be incorporated into the air inlet 112a (Figure 1 A) of the enclosure 104. Though particular example configurations of the heat exchanger 112 are shown in Figure 2A and Figure 2B, in other embodiments, the heat exchanger 112 can also be configured as a plate-and-frame, plate-and-shell, plate fin, or other suitable types of phase-change heat exchanging devices. [0036] As shown in Figure 2A, the enclosure 104 can include a frame 113 to which the heat exchanger 112 is attached via hinges, rivets, or other suitable fasteners. The heat exchanger 112 can include a shell 117 containing multiple tubes 114 extending along a direction that is generally perpendicular to a flow direction of the cooling air 107. The tubes 114 can be constructed from a heat conductive material, such as copper or aluminum and can optionally include fins, baffles, or other suitable heat enhancing components (not shown). Each of the tubes 114 can include an external surface 114a and an internal volume 114b that contains a select amount of the phase change material 116. Though Figure 2A shows the tubes 114 being equally spaced, in other embodiments, the tubes 114 can be spaced in other suitable manners.

[0037] As shown in Figure 2B, the enclosure 104 can have an air inlet 104a, an air outlet 104b, and an internal space 104c between the air inlet 104a and air outlet 104b for housing the computing units 101. The heat exchanger 112 includes a HE inlet 112a and a HE outlet 112b. The HE inlet 112a is in fluid communication with the cool air plenum 105a (Figure 1A). The HE outlet 112b is in fluid communication with the internal space 104c that houses the computing units 101. Though not shown in Figure 2B, the heat exchanger 112 can be attached to the enclosure 104 via hinges, bolts, hangers, or other suitable attachment mechanisms.

100381 During normal operation, cooling air 107 at a temperature lower than the melting point of the phase change material 116 can pass through the gaps 115 between neighboring tubes 114 without melting the phase change material 116 inside the tubes 114. During a cooling outage, the cooling air 107 entering the gaps 115 may be at an elevated temperature that is equal to or higher than the melting point of the phase change material 116 in the tubes 115. As such, heat can be transferred from the incoming air to the phase change material 116 in the tubes 115 and resulting in melting the phase change material 116 while lowering the temperature of the incoming air 107 by, for instance, 5°C to 10°C. As such, the cooling air 107 entering the enclosure 104 can be maintained at a cooling temperature (e.g., 25°C) for a period of time until the air handling units 106 are restarted by, for instance, correcting a chiller failure, a control valve malfunction, a strainer being plugged, or other failures.

[0039] In the illustrated example in Figure 2B, the multiple tubes 115 with the phase change material 116 are arranged in multiple rows 118 with tubes 116 that are staggered with respect to one another. In other examples, the rows 118 can be aligned, interleaved, or have other suitable arrangements to support uniform or differentiated melting of the phase change material 116 along length-, width-, and/or depth-wise planes. In certain embodiments, each row 118 can have tubes 114 that contain different phase change materials 116. For example, a first row 118’ can have tubes 114 that contain a first phase change material having a melting point that is different (i.e., either higher or lower) than a melting point of a second phase change material 116 contained in tubes 114 of a second row 118” downstream of the first row 118. As such, when the second phase change material has a melting point lower than the first phase change material, the incoming air 107 may be cooled to a desired temperature approximating the melting point of the second phase change material. In other examples, the different rows 118 can be arranged to have other suitable profiles of the melting points of the phase change material(s) 116.

[Q04Q] Figures 3A-3C are schematic diagrams illustrating certain components and configurations of a source 120 of cooling fluid suitable for the computing facility in Figure 1 A in accordance with embodiments of the disclosed technology. As shown in Figure 3 A, the source 120 can include a cooling tower 122 operatively coupled to a circulation pump 132 as an adiabatic cooling unit. The cooling tower 122 can include a spray manifold 126 configured to receive cooling fluid return from, for instance, the air handling units 106 in Figure 1 A and a fan 124 that is configured to force air up toward top of the cooling tower 122. The cooling tower 122 can also include a fluid basin 128 configured to contain an amount of cooling fluid (e.g., cooling water) to be provided to a suction of the circulation pump 132. During operation, the cooling fluid return can be sprayed into the cooling tower via the spray manifold 126. Air flowing up by action of the fan 124 can then evaporate a portion of the cooling fluid to reduce a temperature of the cooling fluid. The cooling fluid can then be collected by the fluid basin 128 before being circulated to the air handling units 106 by the circulation pump 132.

[0041 j Figure 3B illustrates another example source 120 for providing cooling fluid to the air handling units 106. As shown in Figure 3B, the example source 120 can be generally similar to that shown in Figure 3A except the source 120 can include a chiller 134 (e.g., a refrigeration unit) that is configured to further reduce a temperature of the cooling fluid 130 from the fluid basin 128. In yet a further example, the example source 120 shown in Figure 3C does not include a cooling water tower 122 (Figure 3B) but instead a fluid reservoir 136 containing the cooling fluid 130 and in fluid communication with the circulation pump 132 and the chiller 134. Though particular examples of the source 120 are shown in Figures 3A-3C, in other embodiments, the source 120 can also include valves, swamp coolers, and/or other suitable devices in addition to or in lieu of those shown in Figures 3A-3C.

[0042] From the foregoing, it will be appreciated that specific embodiments of the disclosure have been described herein for purposes of illustration, but that various modifications may be made without deviating from the disclosure. In addition, many of the elements of one embodiment may be combined with other embodiments in addition to or in lieu of the elements of the other embodiments. Accordingly, the technology is not limited except as by the appended claims.