ACCESS TO SERVICES PROVIDED BY A DISTRIBUTED DATA STORAGE SYSTEM

Description

[0001] The present invention generally relates to communication networks and systems, in particular mobile communication networks and systems.

[0002] Descriptions of such networks and systems can be found in the literature, such as in particular in Technical Specifications published by standardization bodies such as for example 3GPP (3 rd< Generation Partnership Project).

[0003] Documents US 2015 / 117430 A1, US 7 843 843 B1, US 2016 / 1833168 A1 provide technological background.

[0004] In general, in such systems, a terminal (also called User Equipment UE) has access to a network providing communication services. The network generally comprises a Core Network accessed via an Access Network.

[0005] Different systems of different generations have been standardized over time, for example EPS (Evolved Packet System) for 4G generation.

[0006] A Next Generation system (also called 5G), is currently being standardized. An architecture for such Next Generation system is currently being studied, in particular in 3GPP TR 23.799.

[0007] In general, a network comprises various network entities also called Network Functions. For example, in 3GPP TR 23.799, a Network Function is defined as a 3GPP adopted or 3GPP defined processing function in a network, which has defined functional behaviour and 3GPP defined interfaces; a Network Function can be implemented either as a network element on a dedicated hardware, or as a software instance running on a dedicated hardware, or as a virtualised function instantiated on an appropriate platform, e.g. on a cloud infrastructure.

[0008] In 3GPP TR 23.799, a solution for interaction of control plane Network Functions (e.g. Core Control Function CCF) via a Data Layer has been proposed, whereby a Network Function can create, update, delete data in the Data Layer, and other Network Functions can have access to these data, or subscribe to notification upon change of these data, via the Data Layer. A simplified model for interconnection of Core Control Functions via a Data Layer is recalled in Figure 1 taken from 3GPP TR 23.799. The stored data may include context data (e.g. Mobility Management context, Session Management context,...etc.).

[0009] Such Data Layer thus brings several advantages, in particular it supports flexible interconnection between control plane network functions, e.g. avoiding functional and signaling impact to unrelated network functions for a given interaction.

[0010] As recognized by the inventors and as will be described with more detail, there is a need for an optimized solution for access to services provided by such Data Layer, particularly in view of the high level of performances required for Next Generation (or 5G) systems. More generally there is a need to improve such systems, such as Next Generation (or 5G) systems.

[0011] As solution, the present invention provides subject-matter according to the independent claims, preferred embodiments thereof being defined in dependent claims.

[0012] Some embodiments of apparatus and / or methods in accordance with embodiments of the present invention are now described, by way of example only, and with reference to the accompanying drawings, in which: Figure 1 is intended to recall a simplified model for Data Layer interconnection model for Next Generation system architecture, Figure 2 is intended to illustrate in a simplified way an example of workflow related to a first procedure for access to services provided by a Data Layer, according to embodiments of the invention, Figure 3 is intended to illustrate in a simplified way an example of workflow related to a second procedure for access to services provided by a Data Layer, according to embodiments of the invention, Figure 4 is intended to illustrate in a simplified way an example of workflow related to a third procedure for access to services provided by a Data Layer, according to embodiments of the invention. Description of various aspects and / or embodiments of the invention

[0013] As indicated above, 3GPP is currently standardizing the so called Next Generation Core Network which will be the core network for 5G networks. 3GPP SA2 is documenting the results in TR 23.799. In this framework it has been proposed that data (which may include subscriber, session, mobility, policy, operations data) in the 5G core networks is stored in data plane - the so called Shared Data Layer (SDL).

[0014] The SDL enables a flexible cloud-optimized network design with simplified stateless VNFs (Virtualized Network Functions), accessing the Shared Data Layer for creating, reading, updating, deleting data or for being notified about data changes.

[0015] The SDL can be seen as a distributed data storage system in a Cloud for Telco environment, also including a service of notification to applications upon modification of data.

[0016] Although embodiments of the invention will be described more particularly for the SDL case, embodiments of the invention are not limited to the SDL case, and generally apply to access to services provided by a distributed data storage system.

[0017] Access to services provided by a distributed data storage such as the SDL may involve: entities referred to hereafter as client application (or client, or client VNF as illustrated in figures 2 to 4), accessing to these services, entities referred to hereafter as service endpoint (or endpoint, or access point), via which client applications access to these services. The service endpoint refers to the connection to a node within the SDL that is providing access to storage or notification service. It can be considered as a proxy to the Data Base Storage node.

[0018] Client applications may communicate with service endpoints at application level using protocols such as LDAP, RESP, CQL, SOAP....

[0019] Client applications may store their data in one or multiple local or remote storage nodes in the SDL.

[0020] Current state of the art fails to provide an optimized solution for access to services provided by the SDL, at least for the following reasons : No solution is currently standardized (e.g. in the standardized 3GPP User Data Convergence architecture) to allow client applications to discover the available end points for storage and notifications services, and be updated when new end-points are added or existing end-points are removed, as function of dynamic network topology (e.g. scale-in / scale-out operations in the cloud). Besides, existing solutions do not permit to take in account the telco constraint for 5G and reliability, e.g. : ▪ Low latency between the client and the storage to meet the ultra low latency requirement in 5G environment. ▪ Redundancy level to avoid / limit impact on traffic in case of site crash of access point, site or lost of access to a region. ▪ The traffic level requested by the client. In existing solutions, scale-in / scale-out service is not offered : need external application like service discovery to update DNS records. In case of : Scale in : Clients will be only notified when connection will be closed Scale out : only new client that will access to DNS will take benefit of the new configuration.

[0021] Scale-in may be defined as an operation (manual or automatic following metrics analysis) that suppresses an endpoint from the SDL configuration.

[0022] Scale-out may be defined as an operation (manual or automatic following metrics analysis) that adds an endpoint to the SDL configuration.

[0023] Embodiments of the present invention in particular allow to overcome such deficiencies.

[0024] In some embodiments, it is proposed to allow one or more of: ▪ allow client applications of SDL to discover (i.e. be assigned by SDL) end-point accesses to storage service and notification service, taking into account load distribution in the SDL and specific application performance constraints (e.g. latency constraint for 5G). ▪ allow client applications to request the SDL to provide redundancy of their data across multiple local and / or remote access nodes; ▪ enable the SDL to inform client applications of the need of reconfiguration (e.g. addition or removal of end-point accesses) in function of the dynamically evolving network topology (e.g. scale-in / scale-out operations in the cloud) and the load evolution within the SDL (e.g. overload situation) taking in account the high constraint of reliability in telco environment.

[0025] In some embodiments, new procedures between client applications and the SDL are proposed to be introduced.

[0026] In some embodiments: The Shared Data Layer (SDL) may comprise an entity (also called hereafter Service Discovery Function, or Service Discovery, or VNFC Service Discovery as illustrated in figures 2 to 4) providing a service discovery functionality for access to services (storage and notification services) provided by the SDL. The Service Discovery may allows a client application to register to the SDL to discover the endpoints within the SDL supporting a particular service (e.g. data storage or subscription to notification service).

[0027] In such registration procedure, the client may provide information regarding one or more of: its localization, its traffic level, and data redundancy requirements, that allow the Service Discovery of the SDL to assign the most optimal access points to the application, balancing the load on the different access points and taking in account the application latency and redundancy requirements.

[0028] In some embodiments, a Network Function of the 5G network may indicate whether it wants to access the data locally and remotely. So, this allows data (e.g. context data) to be exchanged across Network Functions irrespective of whether they are located close to each other or remotely from each other.

[0029] In addition the SDL Service Discovery may enable the SDL to update the client applications when there is a need for reconfiguration of the access points assigned to the applications.

[0030] In some embodiments: The client may send to the Service Discovery information related to one or more of: its localization, the traffic level requested and the redundancy level.

[0031] The SDL Service Discovery may leverage this information in one or more of the following ways: The localization may permit to provide access point close to the client application and thus a response time in line with the response time constraint of the application, e.g. for 5G. The traffic level requested may permit to estimate the number of access point in the database that must be allocated to the client and to better share the load of all the application clients across the SDL access points. This may also be used to anticipate the triggering of scale-out operations. the data access level (local and remote).

[0032] In some embodiments: The Service Discovery functionality may permit at least one of: to alert the client that a VM will be shut down and provide new access point. To notify the client in case of required rebalancing due to exceed load.

[0033] Examples of new procedures introduced according to embodiments of the invention, are illustrated in a simplified way in figures 2, 3, 4.1) Workflow for client registration and SDL endpoint discovery

[0034] Part or all of following steps (illustrated in figure 2) may be provided: 1- Client may register to Service Discovery. Client may send a Register request including one or more of following information or client characteristics: identity, Client type (e.g. Network Function type, such as HSS, MME, ...etc.), locality , requested endpoints description (e.g. protocol used), traffic qualification (e.g. real-time, provisioning, ...etc.), local and remote strategy redundancy, ...etc. Service Discovery may memorize such Client information. 2- Service Discovery may allocate client UUID (Universal Unique Identifier), and may calculate best endpoints allocation taking into account latency (optimize response time, strong requirement 5G deployment), redundancy model (reliability) and endpoint load 3 - Service Discovery may return UUID, to be used for further Client requests, 4 - Client may request allocated endpoints 5- Service Discovery may return the list of allocated endpoints with connection data (e.g. IP address, port, protocol) for each of them, e.g. endpoints i and j 6 & 7: client may connect to allocated endpoints 2) Workflow for endpoint scale-in: alert of shut down to avoid impact on traffic (smooth operation)

[0035] Part or all of following steps (illustrated in figure 3) may be provided: 1- Service Discovery may be aware of topology change with endpoint i scale-in and may re-calculate best endpoints allocation taking into account endpoint i scale-in, latency, redundancy model and endpoint load and Client information memorized at client registration. E.g., Endpoint k may be allocated to Client 2- Service Discovery may notify Client of topology change in anticipation before the endpoint is suppressed to avoid impacts on QoS (Quality of Service) 3-Client may request allocated endpoints 4- Service Discovery may return the list of allocated endpoints with connection data for each of them 5- Client may connect to new endpoint k before disconnecting from endpoint i to avoid impacts on QoS 6-Client may disconnect from endpoint i 7- endpoint i can be suppressed from SDL configuration 3) Workflow for endpoint scale-out: redistribution of traffic

[0036] Part or all of following steps (illustrated in figure 4) may be provided: 1- Following scale out due to extension capacity (overload / optimization), Service Discovery may be aware of topology change, e.g. with endpoint k scale-out, and may re-calculate best endpoints allocation taking into account new endpoint k, latency, redundancy model and endpoint load. E.g., endpoint k may be allocated to client in addition to already allocated endpoints, to avoid impacts on QoS for that Client. 2- Service Discovery may notify Client of topology change 3 - Client may request allocated endpoints 4- Service Discovery may return the list of allocated endpoints with new endpoint k 5 - Client may redistribute traffic and connect to the new endpoint k, in addition to connections on already allocated endpoints.

[0037] A person of skill in the art would readily recognize that steps of various above-described methods can be performed by programmed computers. Herein, some embodiments are also intended to cover program storage devices, e.g., digital data storage media, which are machine or computer readable and encode machine-executable or computer-executable programs of instructions, wherein said instructions perform some or all of the steps of said above-described methods. The program storage devices may be, e.g., digital memories, magnetic storage media such as a magnetic disks and magnetic tapes, hard drives, or optically readable digital data storage media. The embodiments are also intended to cover computers programmed to perform said steps of the above-described methods.

Claims

1. An apparatus comprising: at least one processor; and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to: - provide functions of a core network control plane function, said functions including access to services provided by a distributed data storage system in a communication core network, said services including at least one of create data, read data, update data, delete data, subscribe to notification of data changes, - acting as a client application, send to a service discovery function a request for access point allocation of access points for said access to services, at least one access point allocated to said client application by the service discovery function based on at least one of: client application requirements, and conditions at the distributed data storage system-.

2. An apparatus according to claim 1, wherein: - client application requirements include at least one of: • location of the client application, • performances required by the client application, • access redundancy level required by the client application.

3. An apparatus according to any of claims 1 or 2, wherein: - conditions at the distributed data storage system include at least one of: • load conditions, • configuration conditions.

4. An apparatus according to any of claims 1 to 3, configured to: - indicate to the service discovery function, if data needs to be stored locally and / or if data needs to be stored remotely.

5. An apparatus according to any of claims 1 to 4, configured to: - indicate its requirements to the service discovery function during a procedure for registration for access point discovery.

6. An apparatus according to any of claims 1 to 5, configured to: - provide in a request for access point allocation sent to the service discovery function, identification information received from the service discovery function during a procedure for registration for access point discovery.

7. An apparatus according to any of claims 1 to6, configured to perform at least one of: - receive from the service discovery function a notification in anticipation before an already allocated access point is suppressed, - upon reception of a notification from the service discovery function, send to the service discovery function a new request for access point allocation, - connect to a new allocated access point before disconnecting from an already allocated access point.

8. An apparatus according to any of claims 1 to 7, configured to perform at least one of: - receive from the service discovery function a notification indicating need of allocation of a new access point in addition to already allocated access points, - upon reception of a notification from the service discover function, send to the service discovery function a new request for access point allocation, - connect to a new allocated access point, in addition to connection to already allocated access points.

9. An apparatus comprising: at least one processor; and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to: provide functions of a service discovery function, said functions including: receive from a core network control plane function acting as a client application, a request for access point allocation for access points for access to services provided by a distributed data storage system in a communication core network, said services including at least one of create data, read data, update data, delete data, subscribe to notification of data changes, - allocate to the client application at least one access point for said access, based on at least one of: client application requirements, and conditions at the distributed data storage system.

10. An apparatus according to claim 9, wherein: - client application requirements include at least one of: • location of the client application, • performances required by the client application, • data redundancy level required by the client application.

11. An apparatus according to any of claims 9 or 10, wherein: - conditions at the distributed data storage system include at least one of: • load conditions, • configuration conditions.

12. An apparatus according to any of claims 9 to 11, configured to: - receive from the client application an indication if data needs to be stored locally and / or if data needs to be stored remotely.

13. An apparatus according to any of claims 9 to 12, configured to: - receive client application requirements during a procedure for registration for access point discovery.

14. An apparatus according to any of claims 9 to 13, configured to: - memorize information received from the client application during a procedure for registration for access point discovery.

15. An apparatus according to any of claims 9 to 14, configured to: - receive from the client application a request for access point allocation.

16. An apparatus according to any of claims 9 to 15, configured to: - receive from the client application, in a request for access point allocation, identification information sent to the client application during a procedure for registration for access point discovery.

17. An apparatus according to any of claims 9 to 16, configured to perform at least one of: - send to the client application a notification in anticipation before an already allocated access point is suppressed, - receive from the client application a new request for access point allocation, - allocate at least one new access point, upon new request for access point allocation from the client application.

18. An apparatus according to any of claims 9 to 17, configured to perform at least one of: - send to the client application a notification indicating need of allocation of a new access point in addition to already allocated access points, - receive from the client application a new request for access point allocation, - allocate at least one new access point, upon new request for access point allocation from the client application.

19. A methodby an apparatus, the method comprising: providing functions of a core network control plane function, said functions including access to services provided by a distributed data storage system in a communication core network, said services including at least one of create data, read data, update data, delete data, subscribe to notification of data changes; and acting as a client application, sending to a discovery function a request for access point allocation of access points for said access, at least one access point allocated to said client application by the service discovery function based on at least one of: client application requirements, and conditions at the distributed data storage system.

20. A method fby an apparatus, the method comprising: providing functions of a service discovery function, said functions including: receiving from a core network control plane function acting as a client application, a request for access point allocation of access points for access to services provided by a distributed data storage system in a communication core network, said services including at least one of create data, read data, update data, delete data, subscribe to notification of data changes; and allocating at least one access point for said access, to the client application, based on at least one of: client application requirements, and conditions at the distributed data storage system.