Background

[0001] Network Functions Virrualization (NFV) is an emerging design approach for constructing Information Technology (IT) applications, particularly in the telecommunications industry. The classical approach to network architecture is based upon fragmented, purpose built hardware for implementing network functions - also known as Physical Network Functions (PNFs) (e.g. firewalls, Deep Packet Inspectors (DPI)), Network Address Translators (NAT)) which require physical installation at every site at which they are needed. In contrast, NFV aims to consolidate many network equipment types onto, for example, standardized high volume servers, switches, and storage through the implementation of Virtualized Network Functions (VNFs) in software which can run on a range of standard hardware. Furthermore, NFV aims to transform network operations because the VNFs can be dynamically moved to, or instantiated in, various locations in the network as required without the need for installation of new hardware.

Allocation of resources to a VNF is typically made via an NFV Orchestrator (NFVO), which includes the system overall resource management view. In cases where the NFVO is bypassed in resource allocation, the consistency of the overall resource management view can be compromised, reducing the efficiency of the system resulting in resources being lost. It is ail object of the implementations described herein to mitigate against at least some of these problems.

Brief Description of the Drawings

[0002] Various features and advantages of the present disclosure will be apparent from the detailed description which follows, taken in conjunction with the accompanying drawings, which together illustrate, by way of example only, features of the present disclosure, and wherein:

[0003] Figure 1 shows a simplified schematic representation of an exemplary architecture illustrating an exemplary implementation;

[0004] Figure 2 is a flow diagram illustrating a method according to an exemplary implementation;

[0005] Figure 3 shows a simplified schematic representation of an exemplary architecture illustrating an exemplary implementation;

I [0006] Figure 4 is a flow diagram illustrating a method according to an exemplary implementation;

[0007] Figure 5 is a flow diagram illustrating a method according to an example;

[0008] Figure 6 is a flow diagram illustrating a method according to an example;

[0009] Figure 7 illustrates a server; and

[0010] Figure 8 illustrates a server.

Detailed Description

[0011 ] The following abbreviations which may be found in the specification and/or the drawings are defined as follows:

IM Infrastructure Manager

NFV Network Functions Virtual! zation

NFVI NFV Infrastructure

NFVO NFV Orchestrator

OSS Operation Support Systems

PNF Physical Network Function

VIM Virtual IM

VI Virtual Infrastructure

VNF Virtual Network Function

VNFM Virtual Network Function Manager

[0012] The European Telecommunications Standards Institute (ETSI) define an architecture for NFV in the standard "ETSI NFV Reference Architecture Framework ETSI GS NFV 002 VI .1.1 (2013-10)". In this architecture, the NFV Infrastructure (NFVI), comprising hardware resources such as servers, storage, and networking components, as well as virrualized computing, storage, and networking, is managed by a Virtuaiized Infrastructure Manager (VIM). The VNFs utilizing resources of the NFVI are managed by VNF Managers (VNFM). Operation of the IM(s) and the VNFM(s) are overseen by an NFV Orchestrator (NFVO), which includes the system overall resource management view across multiple instances of the NFV infrastructures. In this reference architecture, the NFV Orchestrator (NFVO) interfaces directly with the VNFMs and the VIMs.

[0013] In a normal operation of the NFV architecture, as defined in ETSI standard use cases, when a VNF requires a resource, the associated. VNFM makes a request to the NFVO to allocate infrastructure resources. This request may comprise, for example, a request for a Virtual Machine with W GB of Random Access Memory, XGB of storage, Operating System Y, and Virtual Local Area Network connection Z. In response, the NFVO selects the most appropriate available resource(s). The response may comprise, for example, the allocated Virtual Machine Identifier, a login name, and security credentials for use of the resource. The NFVO prepares and provisions the selected resource(s) via the VIM. Once the resource(s) is/are ready for use, the NFVO informs the VNFM of the allocated resource(s). In allocating resources in such a way, at all times the NFVO has an exact view of what infrastructure is in use, and to whom it is allocated,

[0014] In some scenarios however, elastic operation is required and/or desired, and so a VNFM may need to make a request for more or less resources. This can be effected using a similar procedure as described above, namely making a request directly to the NFVO, who then may allocate more or less resources as appropriate. This procedure ensures the NFVO maintains a complete and accurate overview of the allocated infrastructure.

[0015] An alternative scenario, also supported by ETSI use cases, allows the VNFN to request directly resources from the VIM, bypassing the NFVO. Bypassing the NFVO however may lead to a compromise of the completeness and accuracy of the NFVO's overview of allocated infrastructure. Even if the VNFM informs the NFVO of the change of allocation, any delays in the communication of this information may lead to inconsistencies in the allocation data at the NFVO, and so may reduce the effectiveness and efficiency of the system. Furthermore, using this approach, if there are any failures in the VNFM, then the associated resources may become orphaned from the system and effectively lost. The NFVO will no longer have an accurate view of the allocated and free resources.

[0016] Referring now to the drawings, and first to Figure 1, there will be described an NFV architecture.

[0017] Figure 1 shows a simplified schematic representation of an NFV system architecture. The system architecture comprises a NFVO 102.

[0018] The NFVO 102 is in direct communication with Operation Support Systems (OSS) 104 which are used at a high level to administer and maintain network systems. The NFVO 102 may, for example, ran as software on a physical or virtual server (not shown) to operate, manage and automate the software and infrastructure associated with NFV resources in order to implement, for example, a defined service model.

[0019] The NFVO 102 is also in direct communication with at least one VNFM 1 10 which may, for example, ran as software on a physical or virtual server (not shown), via interface Or-Vnfm 105. The VNFM 1 10 manages, via interface Ve-Vnfm 103, instances of virtual network functions (VNFs 108a to 108c) used, for example, by various applications (APPs 106a to 106c), for example Traffic Steering applications, running on the network. VNFs 108a to 108c are implementations of a network function that can be deployed on a suitable NFV Infrastructure (NFVI) 122 (i.e. the totality of all hardware and software components that build up the environment in which VNFs are deployed), typically running as a virtual machine in a Hypervisor (or other resource pooling/sharing) environment.

[0020] The NFVO 102 is also in direct communication with Virtualized Infrastructure Manager VIM 120, which may, for example, ran as software on a physical or virtual server (not shown), via interface Or-Vi 107. The VIM 120 manages, via interface Nf-Vi 1 11, the NFVI 122 used, for example, by the VNFs 108a to 108c to perform their function. The NFVI 122 may comprise Virtual Infrastructure (VI) 1 16, for example, virtual computing, storage, and. networking infrastructure, for example, hardware resources such as physical computing and storage resources and a physical communications network. The VIM may comprise a component 118 for processing and/or responding to resource allocation requests. In the standard. ETSI Reference Architecture Framework, the VNFM 1 10 is in direct communication with the VIM 120 via interface Vi-Vnfm (not shown in figure 1).

[0021 ] As described above, in the normal operation of the NFV architecture, as defined in ETSI standard use cases, when a VNF 108a to 108c requires a resource, for example a physical storage resource, the associated VNFM 110 makes a request to the NFVO 102 to allocate the infrastructure resource. In response, the NFVO 102 selects the most appropriate available resource, in this example a physical storage resource, from a database of resources 113 of the NFVI 122. The NFVO 102 then communicates with the VIM 120 to prepare and provision the selected resource of the NFVI 122. Once it as been communicated by the VIM 120 to the NFVO 102 that the resource has been prepared and provisioned and. is ready for use, the NFVO 102 updates its resource allocation database 1 13, and informs the VNFM 110 of the allocated resource. The VNF 108a to 108c associated with the initial request for resources is then free to use the allocated resource. As described above, in allocating resources in such a way, the NFVO 102 has an exact view of what infrastructure is in use, and to whom it is allocated.

[0022] In an alternative operation of the architecture of Figure 1 , the VNFM 1 10 may bypass the NFVO 102 in requesting additional or reduced resources of the NFVI 122 by communicating with the VIM 120 directly. Bypassing the NFVO 102, however, may lead to a compromise of the completeness and accuracy of the NFVQ's overview (e.g. the allocation database 113 of the NFVO 102) of allocated infrastructure. Even if the VNFM 1 10 were to inform the NFVO 102 of the change of allocation of the resource of NFVI 122, any delays in the communication of this information may lead to inconsistencies in the allocation data at the NFVO 102, and so may reduce the effectiveness and efficiency of the system. Furthermore, using this alternative operation, if there are any failures in the VNFM 110, then the associated resources of the NFVI 122 may be orphaned from the system and effectively lost from the NFVO's 102 overview.

[0023] Exemplary implementations allowing a VNFM 1 10 to request and be allocated resources directly from a VIM 120 whilst ensuring the maintenance of the NFVO's 102 overview involve a proxy in the communications path between the at least one VNFM 1 10 and VIM 120, which are described below with reference to Figures 1 and 2.

[0024] In Figure I , in in the communications path between the at least one VNFM 1 10 and VIM 120 there is a proxy server 200, and the proxy server 200 is in direct communication with the NFVO 102.

[0025] In some exemplary implementations, the proxy 200 may be made thin in the network, i.e. the delay caused, by the operations of the proxy 200 are near negligible in the network.

[0026] There is an existing interface Vi-Vnfm, between the VNFM 1 10 and the VIM 120 in the standard ETSI Reference Architecture Framework, in which the proxy 200 may be placed as shown in Figure 1. However, to allow direct communication between the proxy 200 and the NFVO 102, a new interface 1 15 not part of the current ETSI Reference Architecture Framework is introduced.

[0027] The purpose of the proxy 200 is to ensure that an accurate record of the allocated and free resources inside an NFV deployment is maintained at the NFVO 102, without limiting the mode of operation of the VNFM 110 to requesting and receiving resource allocation through the NFVO 102. This is advantageous in both enabling flexibility and ensuring consistency of resource allocation in a multi-vendor environment.

[0028] An exemplary mode of operation of the proxy 200 in the exemplary implementation of Figure 1 will now be described in more detail with reference to Figure 2, which is a flow diagram showing a method according to an exemplary implementation.

[0029] In an exemplary mode of operation, in the architecture shown in Figure 1, a VNFM 1 10 makes a request to the VIM 120 for more or less resources of the NFVI 122. In S302 of Figure 2, the request is intercepted by the proxy 200 and the contents read. The request is then forwarded to the VIM 120 and also sent in near real time to the NFVO 102. In S304, a response from the VIM 120, for example, allocating the requested resources, is intercepted by the proxy 200 and forwarded to the VNFM 1 10 and also sent in near real time to the NFVO 102. The information in the request and/or the response may be used, for example, by a resource manager of the NFVO 102 to validate the requests and/or update a resource allocation database 1 13. In some exemplary implementations, the proxy 200 may wait until it has received the response from the VIM 120, before sending both the request and the response to the NFVO 102 simultaneously or near simultaneously.

[0030] The database 1 13 is a database of all of the resources of the NFVI 122 and of authorized resource allocations, for example, resources authorized for an individual sendee provider, for each APP 106a to 106c, or for each VNF 108a to 108c. In S308, information relating to an intercepted request and/or response is compared to the database 1 13. In S310, a determination is made whether the request and/or response would result in an authorized allocation of resources or not. If it is determined that the request and/or response would result in an authorized allocation of resources, in S312 the NFVO 102 may record the request data and the subsequently intercepted response data, and update the resource allocation database 1 13 accordingly. If, however, it is determined by the NFVO 102 that the request and/or response would result in an unauthorized allocation of resources, or is otherwise is in error, for example, is a request for a resource that is already allocated, then in S314, the NFVO 102 may raise an alert to a higher level of service management, for example to an Operations Support System, which may intervene. This alert may, for example, trigger a search for and/or resolution of corruption in the resource database, or, for example, an inquiry into the reason why an unauthorized request was sent.

[0031] An alternative exemplar}' implementation is shown in Figure 3, which shows essentially the same architecture of figure 1, except for that the proxy 200 is located within, i.e. is a part of, the VIM 120.

[0032] The different implementations of the proxy 200 of Figure 1 and Figure 3 may have differing implications for the modification of the downstream trust chain from the VNFM 110 to the VIM 120 due to the proxy 200. In the case of the 'standalone' proxy 200 of Figure i , the proxy may not be totally invisible. In the case of the proxy 200 incorporated into the VIM 120 of Figure 3, there will be no impact on the trust chain. For example, the standalone proxy 200 may have its own identity and security credentials, whereas this may not be necessary when incorporated to the VIM. An exemplary mode of operation in the architecture shown in Figure 4 is the same as that described above with reference to the architecture of Figure 1 and Figure 2, except that since the proxy 200 in the architecture shown in Figure 3 is within, i.e. is a part of the VIM 120, rather than requests and responses between the VNFM 1 10 and the VIM 120 being intercepted and forwarded by the proxy 200, such requests and responses are instead received and sent by a proxy function of the VIM 120 to a component 1 18 of the VIM 120. This component may be, for example a component of the VIM 120 for processing and/or responding to resource allocation requests.

[0033] The use of the proxy 200 in these exemplary modes of operation minimizes the time the resource allocation database 1 13 of the NFVO 102 is not up to date. For a request from a VNFM 1 10 that is authorized, the database 1 13 will be updated shortly after the proxy 200 informs the NFVO 102 of the response from the VIM 120. This is as compared to a situation as described above in the standard ETSI Reference Architecture Framework, where the NFVO 102 is informed by the VNFM 1 10 of a resource allocation only after the VNFM 110 receives all the responses from the VIM 120 informing it of the complete resource allocation for this VNF. Further, in the situation as in the reference architecture, if, for example, a VNFM 110 tails after it has been allocated resources, but before it has reported the allocation to the NFVO 102, the allocation state of the resources may be inconsistent with the allocation database 1 13 of the NFVO 102. This may lead to unexpected behaviors of the automation of the allocation of resources by the NFVO 102, and may lead to the loss of resources from the system, and/or major system failures that would be hard to recover from without reinitializing the system. The use of the proxy 200 in these exemplary modes of operation, however, ensures that a response from a VIM 120 allocating resources to a VNFM 110 is intercepted for forwarding to the NFVO 102 for updating its allocation database 1 13 before the VNFM 1 10 receives the response. Hence even if the VNFM 1 10 fails, the overview of the allocation of resources at the NFVO 102 is up to date.

[0034] The proxy 200 therefore ensures that an accurate record of the allocated and free resources inside an NFV deployment may be maintained, without limiting that a request for resources by a VNFM 1 10 must go via an NF VO 102. Regardless of whether the NF VO 102 is bypassed or not, the proxy 200 therefore ensures there is a consistent overview of resource allocation, ensuring the maximum utilization of resources and reducing management costs.

[0035] In some exemplary implementations, the proxy 200 may be made active, as described below with reference to Figure 4.

[0036] The proxy 200 may be configured by the NFVO 102, for example by a resource manager of the NFVO 102, to only allow valid requests to pass from the VNFM 1 10 to the VIM 120 (or one of its components). For example, the NFVO 102 may communicate to the proxy 200 which sources, for example, which VNFs 108a to 108c, APPs 106a to 106c, or VNFM 1 10 are authorized to be allocated which resource. In S502 of Figure 4, the proxy 200 intercepts (or alternatively receives) and reads a request from a VNFM 1 10 to a VIM 120. [0037] In 8504, the proxy 200 compares the source and resources of the request to authorized resource allocations. In S506, on the basis of the comparing, the proxy 200 decides whether the request is valid, for example if fulfilment of the request would result in an authorized allocation of resources or not. If the proxy 200 intercepts (or receives) a request from a VNFM 110 for resources for which there is no authorization, then, in S508, the proxy 200 does not forward (or pass on) the request to the VIM 120 (or one of its components). The proxy 200 may alert the NFVO 102 to the fact that such an unauthorized, request has been made, which may be passed on by the NFVO 102 to a higher level management. The proxy 200 may, for example, in response to receiving a unauthorized request, be configured to generate a response to the unauthorized request informing the VNFM 110 that the request has been denied. The VNFM 1 10 may then resend its request to the NFV 102 in order to be allocated resources which have been authorized for the VNFM 1 10 to use. If the request is determined in S506 to be valid however, then in 8510 the proxy 200 forwards (or passes on) the request to the VIM 120 (or one of its components) and also sends the request to the NFVO 102. In 8512, the proxy 200 may then intercept, read and forward the response from the VIM 120 (or one of its components) to the VNFM 1 10, and send the response to the NFVO 102. In S516 the NFVO 102 may record the received request and response data and, for example, update the resource allocation database 1 13 accordingly.

0038] The proxy 200 may also be configured by the NFVO 102, for example by a resource manager of the NFVO 102, to rewrite a request received from a VNFM 110 before forwarding or passing on the request to VIM 120 (or one of its components). For example, the NFVO 102 may communicate to the proxy 200 that a request for resources which are unauthorized be rewritten as a request only for those resources of the original request which are authorized for use. As a further example, during times of service migration, during system recover}', or at times of peak demand, the NFVO 102 may communicate to the proxy 200 to rewrite a request for resources so that the request is directed to a different VIM 120 than the one specified in the original request. In a still further example, the NFVO may communicate to the proxy to implement policy controls, for example rewrite a request for resources so as to limit the resources allocated to lower priority requests.

[0039] The described implementations therefore ensure a maximum use of resources between different VNFs controlled by multiple VNFMs to the extent that they are authorized. This is critical for NFV as it is based on multiple applications running for multiple vendors in the same hosting environments. Further, the exemplary implementations may prevent the wrongful or erroneous allocation of resources, and so reduce the risk of multiple applications failing at the same time arising from the automated nature of the errors,

[0040] Figure 5 is a flow diagram of an exemplary generalized method for operating a proxy 200 of the present invention. In S602, a resource allocation request is received from a VNFM 110. In S604, the resource allocation request is sent to a component of a VIM 120. In S606, a response to the resource allocation request is received from the component of the VIM 120. In S608, the response is sent to the NFVG 102.

[0041 ] Figure 6 is a flo diagram of an exemplar generalized method for operating an NFV Orchestrator 102 of the present invention. In S702, a response to a resource allocation request is received, via interface 1 15 with, a proxy 200, the response originating from a component of a VIM 120 in response to a resource allocation request from a VNF manager 110 forwarded to the component of the VIM by the proxy.

[0042] Figure 7 depicts example components of a server 802, which in one example, implements the proxy 200 of Figures 1 and 3. The server 802 includes software 804, which includes adaptation software 820 that is executable on one or more multiple processors 806 to implement the described functions of the proxy 200.

[0043] A processor 806 can include a microprocessor, microcontroller, processor module or subsystem, programmable integrated circuit, programmable gate array, or another control or computing device.

[0044] The server 802 further comprises one or more network interfaces 808 to allow the server 802 to communicate with the VNFMs 110, NFVOs 102, and, where necessary IMs 120.

[0045] The server 802 further comprises a storage medium 810, which can store various software, data, machine-readable instructions, or any other type of information required for the proxy 200 to implement its functions. The storage medium may store proxy software 804, which may comprise adaptation software 820 for adapting a proxy to implement any of the functions of proxy 200 as described above.

[0046] The storage medium (or storage media) 810 can be implemented as one or multiple computer-readable or machine-readable storage media. The storage media include different forms of memory including semiconductor memory devices such as dynamic or static random access memories (DRAMs or SRAMs), erasable and. programmable read-only memories (EPROMs), electrically erasable and programmable read-only memories (EEFROMs) and flash memories; magnetic disks such as fixed, floppy and removable disks; other magnetic media including tape; optical media such as compact disks (CDs) or digital video disks (DVDs): or other types of storage devices. Note that the instructions discussed above can be provided on one computer-readable or machine-readable storage medium, or alternatively, can be provided on multiple computer-readable or machine-readable storage media distributed in a large system having possibly plural nodes. Such computer-readable or machine-readable storage medium or media is (are) considered to be part of an article (or article of manufacture). An article or article of manufacture can refer to any manufactured single component or multiple components. The storage medium or media can be located either in the machine running the machine-readable instructions, or located at a remote site from which machine-readable instructions can be downloaded over a network for execution.

[0047] Figure 8 depicts example components of a server 902, which in one example, implements the FVO 102 of Figures 1 and3. The server 902 includes software 904, which includes adaptation software 920 that is executable on one or more multiple processors 906 to implement the described functions of the NFVO 102,

[0048] A processor 906 can include a microprocessor, microcontroller, processor module or subsystem, programmable integrated circuit, programmable gate array, or another control or computing device.

[0049] The server 902 further comprises one or more network interfaces 908 to allow the server 902 to communicate with the OSS 104, VNFMs 1 10, proxy 200, and VIMs 120.

[0050] The server 902 further comprises a storage medium 910, which can store various software, data, machine-readable instructions, or any other type of information required for the NFVO 02 to implement its functions. The storage medium may store NFVO software which may comprise adaptation software 920 for adapting a NFVO to implement any of the functions of the NFVO 102 as described above. The siorage medium 910 may comprise the resource allocation database 13,

[0051] The storage medium (or storage media) 910 can be implemented as one or multiple computer-readable or machine-readable storage media. The storage media include different forms of memory including semiconductor memory devices such as dynamic or static random access memories (DRAMs or SRAMs), erasable and programmable read-only memories (EPROMs), electrically erasable and programmable read-only memories (EEPROMs) and flash memories; magnetic disks such as fixed, floppy and removable disks; other magnetic media including tape; optical media such as compact disks (CDs) or digital video disks (DVDs); or other types of storage devices. Note that the instructions discussed above can be provided on one computer-readable or machine-readable storage medium, or alternatively, can be provided on multiple computer-readable or machine-readable storage media distributed in a large system having possibly plural nodes. Such computer-readable or machine-readable storage medium or media is (are) considered to be part of an article (or article of manufacture). An article or article of manufacture can refer to any manufactured single component or multiple components. The storage medium or media can be located either in the machine running the machine-readable instructions, or located at a remote site from which machine-readable instructions can be downloaded over a network for execution.

[0052] The above implementations are to be understood as illustrative examples. Further implementations are envisaged. It is to be understood that any feature described in relation to any one example may be used alone, or in combination with other features described, and may also be used in combination with one or more features of any other of the examples, or any combination of any other of the examples. Furthermore, equivalents and modifications not described above may also be employed without departing trom the scope of the invention, which is defined in the accompanying claims.