System and method for intelligently managing sessions in a mobile network
By introducing an intelligent selection engine into the mobile network to dynamically select appropriate network nodes, the flexibility and load balancing issues of session management in existing technologies are solved, session establishment efficiency and network resource utilization efficiency are improved, and the integrity of billing data is ensured.
Patent Information
- Application Number
- CN201980047749.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2018-07-23
- Filing Date
- 2019-07-23
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2039-07-23
AI Technical Summary
In existing technologies, session management in mobile networks lacks flexibility and effective load balancing mechanisms, resulting in insufficiently intelligent selection of network nodes and an inability to dynamically respond to changes in the capacity and status of network components, thus affecting session establishment latency and the coordination of billing data.
An intelligent selection engine is introduced to analyze subscriber session data, select the peer selector with the highest priority, and choose the most suitable network node based on load level and availability. The DNS server and peer list are dynamically adjusted to achieve load balancing and path management.
It improves session establishment efficiency, reduces latency, ensures the integrity of billing data and optimizes the use of network resources, and enhances network flexibility and scalability.
Smart Images

Figure CN113169988B_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims priority to U.S. Provisional Application No. 62 / 702,093, filed on July 23, 2018, entitled “System and Method for Intelligently Managing Sessions In A Mobile Network,” the contents of which are incorporated herein in their entirety. Technical Field
[0003] Embodiments of the present disclosure relate generally to mobile networks, and more particularly to managing sessions in a mobile network. Background Art
[0004] Wireless operators deploy multiple gateway nodes (e.g., Serving Gateway (SGW) / Packet Data Network Gateway (PGW) / System Architecture Evolution Gateway (SAEGW)) network elements to provide services to millions of subscriber Packet Data Network (PDN)-sessions and multiple Access Point Names (APNs) in their networks (because a single node cannot handle that much capacity). Conventionally, to load balance PDN-sessions across these SGW / PGW / SAEGW network elements, the Mobility Management Engine (MME) uses a simple round-robin algorithm. To load balance PDN-sessions, the MME queries the Domain Name Service (DNS) network element to return a list of SGW / PGW / SAEGW network elements that it can select for a given APN and subscriber tracking area, and then polls for new PDN-sessions across the given list of SGW / PGW / SAEGW network elements. This approach is limited because the MME cleans up a static list that does not take into account changing conditions (e.g., network node capacity, PDN-sessions supported by network nodes, network node status). Summary of the invention
[0005] Systems and methods for intelligently managing sessions in a mobile network are described herein. In some embodiments, a method includes: receiving, by a selection engine, a trigger to select a peer node for a subscriber session; selecting, by the selection engine, a peer selector from one or more available peer selectors in response to the trigger; determining, by the selection engine, whether the selected peer selector is associated with a fully qualified domain name (FQDN); determining, by the selection engine, in response to determining that the selected peer selector is associated with the FQDN, whether a domain name system (DNS) server is configured; requesting, by the selection engine, a peer list associated with the FQDN from the DNS server in response to determining that the DNS server is configured; and selecting, by the selection engine, a peer node from the peer list.
[0006] In some embodiments, the trigger comprises at least one of: an indication of session-related control traffic to traverse nodes associated with the selection engine, or a request to select a peer node.
[0007] In some embodiments, selecting a peer selector also includes: analyzing subscriber session data to determine one or more parameters of the subscriber session; identifying one or more matching peer selectors having a key type that matches the one or more parameters among one or more available peer selectors; and selecting a peer selector from the one or more matching peer selectors based on: the selected peer selector having the highest priority among the one or more matching peer selectors.
[0008] In some embodiments, the DNS server is configured when at least one of the primary DNS server or the secondary DNS server is provisioned and reachable.
[0009] In some embodiments, in response to determining that (a) the selected peer selector is not associated with a FQDN, or (b) the DNS server is not configured, the peer list is obtained by the selection engine from the selected peer selector.
[0010] In some embodiments, selecting a peer node from a peer list also includes: receiving the peer list by a selection engine; excluding one or more unreachable peer nodes from the peer list by the selection engine; excluding one or more overloaded peer nodes from the peer list by the selection engine; grouping the one or more remaining peer nodes in the peer list based on a load level associated with each of the one or more remaining peer nodes in the peer list by the selection engine to generate a plurality of peer groups each including one or more peer nodes; and selecting a peer node from the peer group with the smallest load among the plurality of peer groups by the selection engine.
[0011] In some embodiments, each unreachable peer node of the one or more unreachable peer nodes is at least one of: out of order, under maintenance, or disconnected from the network.
[0012] In some embodiments, each overloaded peer node of the one or more overloaded peer nodes is at least one of: at maximum capacity, within a threshold of reaching maximum capacity, or experiencing issues associated with software corruption.
[0013] In some embodiments, the method further comprises: receiving, by the selection engine, a status message identifying at least one of: one or more unreachable peer nodes or one or more overloaded peer nodes.
[0014] In some embodiments, the peer node selects from the least loaded peer group using at least one of round robin, weighted round robin, or hashing.
[0015] In some embodiments, the selection engine is implemented in one of: a mobility management engine for selecting a serving gateway peer node; an evolved packet data gateway for selecting a packet data network gateway peer node; a control plane node for selecting a user plane peer node; or a GPRS tunneling protocol agent (GTP agent) for selecting a gateway peer node.
[0016] In some embodiments, a system includes: a memory; and one or more processors coupled to the memory, the one or more processors configured to read instructions from the memory, which instructions, during execution, cause the one or more processors to perform operations, the operations including: receiving a trigger for selecting a peer node for a subscriber session; in response to the trigger, selecting a peer selector from one or more available peer selectors; determining whether the selected peer selector is associated with a fully qualified domain name (FQDN); in response to determining that the selected peer selector is associated with the FQDN, determining whether a domain name system (DNS) server is configured; in response to determining that the DNS server is configured, requesting a peer list associated with the FQDN from the DNS server; and selecting a peer node from the peer list by a selection engine.
[0017] In some embodiments, selecting a peer selector also includes: analyzing subscriber session data to determine one or more parameters of the subscriber session; identifying one or more matching peer selectors having a key type that matches the one or more parameters among one or more available peer selectors; and selecting a peer selector from the one or more matching peer selectors based on: the selected peer selector having the highest priority among the one or more matching peer selectors.
[0018] In some embodiments, in response to determining that (a) the selected peer selector is not associated with a FQDN, or (b) the DNS server is not configured, the peer list is obtained by the selection engine from the selected peer selector.
[0019] In some embodiments, selecting a peer node from a peer list also includes: receiving the peer list by a selection engine; excluding one or more unreachable peer nodes from the peer list; excluding one or more overloaded peer nodes from the peer list; grouping one or more remaining peer nodes based on a load level associated with each of the one or more remaining peer nodes in the peer list to generate a plurality of peer groups each including one or more peer nodes; and selecting a peer node from the peer group with the smallest load among the plurality of peer groups.
[0020] In some embodiments, a non-transitory computer-readable medium stores instructions that, when executed by a selection engine comprising one or more hardware processors, cause the selection engine to perform operations including: receiving a trigger to select a peer node for a subscriber session; in response to the trigger, selecting a peer selector from one or more available peer selectors; determining whether the selected peer selector is associated with a fully qualified domain name (FQDN); in response to determining that the selected peer selector is associated with the FQDN, determining whether a domain name system (DNS) server is configured; in response to determining that the DNS server is configured, requesting a peer list associated with the FQDN from the DNS server; and selecting, by the selection engine, a peer node from the peer list.
[0021] In some embodiments, the trigger comprises at least one of: an indication of session traffic traversing nodes associated with the selection engine, or a request to select a peer node.
[0022] In some embodiments, selecting a peer selector also includes: analyzing subscriber session data to determine one or more parameters of the subscriber session; identifying one or more matching peer selectors having a key type that matches the one or more parameters among one or more available peer selectors; and selecting a peer selector from the one or more matching peer selectors based on: the selected peer selector having the highest priority among the one or more matching peer selectors.
[0023] In some embodiments, in response to determining that (a) the selected peer selector is not associated with a FQDN, or (b) the DNS server is not configured, the peer list is obtained by the selection engine from the selected peer selector.
[0024] In some embodiments, selecting a peer node from a peer list further includes: receiving the peer list; excluding one or more unreachable peer nodes from the peer list; excluding one or more overloaded peer nodes from the peer list; grouping the one or more remaining peer nodes based on a load level associated with each of the one or more remaining peer nodes in the peer list to generate a plurality of peer groups each including one or more peer nodes; and selecting a peer node from the peer group with the smallest load among the plurality of peer groups.
[0025] These and other capabilities of the disclosed subject matter will be more fully understood after reviewing the following drawings, detailed description and claims.It is to be understood that the phraseology and terminology employed herein are for the purpose of description and should not be regarded as limiting. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Various objects, features, and advantages of the disclosed subject matter may be more fully understood with reference to the following detailed description of the disclosed subject matter when considered in conjunction with the following drawings, in which like reference numerals identify like elements.
[0027] Figure 1 is a system diagram illustrating a networked system.
[0028] Figure 2 is a system diagram illustrating a networked system according to some embodiments of the present disclosure.
[0029] Figure 3 is a system diagram showing an MME including an intelligent selection engine according to some embodiments of the present disclosure.
[0030] Figure 4 is a system diagram showing an ePDG including an intelligent selection engine according to some embodiments of the present disclosure.
[0031] Figure 5 is a system diagram showing a control plane for a SGW, a PGW or a TDF including an intelligent selection engine according to some embodiments of the present disclosure.
[0032] Figure 6 is a system diagram illustrating an independent intelligent selection engine according to some embodiments of the present disclosure.
[0033] Figure 7 is a system diagram showing a GTP agent including an intelligent selection engine according to some embodiments of the present disclosure.
[0034] Figure 8 is a schematic diagram illustrating a peer selector according to some embodiments of the present disclosure.
[0035] Fig. 9 is a flow chart illustrating a process of selecting a peer list according to some embodiments of the present disclosure.
[0036] Fig.10 is a flow chart illustrating a process of selecting a peer node from a peer list according to some embodiments of the present disclosure.
[0037] Fig.11 is a system diagram illustrating a networking system including a GTP proxy with an intelligent selection engine according to some embodiments of the present disclosure.
[0038] Fig.12 is a system diagram illustrating a networking system including a GTP proxy with an intelligent selection engine according to some embodiments of the present disclosure.
[0039] Fig.13is a schematic diagram showing a GTP proxy positioned between an SGW and a PGW to implement a create session request according to some embodiments of the present disclosure.
[0040] Fig.14 is a schematic diagram showing a GTP proxy positioned between two SGWs and a PGW in order to implement an internal 4G handover with a SGW change according to some embodiments of the present disclosure.
[0041] Fig.15 is a schematic diagram showing a GTP proxy positioned between two SGWs and a PGW to implement 4G to 3G handover according to some embodiments of the present disclosure.
[0042] Fig.16 is a table showing parameters maintained by a GTP agent for switching tunnels when processing GTP-U traffic and GTP-C traffic according to some embodiments of the present disclosure.
[0043] Fig.17 is a flow chart illustrating processing of a create session request according to some embodiments of the present disclosure.
[0044] Fig.18 is a schematic diagram illustrating path management for echo request / response recovery according to some embodiments of the present disclosure.
[0045] Fig.19 is a schematic diagram illustrating path management for PGW failure according to some embodiments of the present disclosure.
[0046] Fig. 20 is a schematic diagram illustrating path management for SGW failure according to some embodiments of the present disclosure.
[0047] Fig.21 is a system diagram illustrating a networked system according to some embodiments of the present disclosure. DETAILED DESCRIPTION
[0048] This article describes a system and method for introducing an intelligent selection engine within a network to solve the problems described above. The intelligent selection engine can be located at multiple points in the network to provide load balancing, path management, tunnel switching, topology hiding, call detail record (CDR) generation, APN name manipulation, Gy interface functionality, and local and geographic redundancy.
[0049] Wireless operators typically deploy multiple Serving Gateways (SGWs), Packet Data Network Gateways (PGWs), and Gateway General Packet Radio Service (GPRS) Support Nodes (GGSNs) network elements in their networks to serve millions of subscribers for multiple Access Point Names (APNs) (because a single node cannot handle that much capacity). When a subscriber's session is established in the network, a Mobile Management Entity (MME) / Serving General Packet Radio Service (GPRS) Support Node (SGSN) uses a Domain Name System (DNS) query to retrieve a list of SGWs / PGWs / GGSNs from which the Mobile Management Entity (MME) / Serving General Packet Radio Service (GPRS) Support Node (SGSN) can select for a given APN. If an operator wishes to add, delete, or move a SGW / PGW / GGSN from its network, corresponding changes are made to the DNS nodes. Making such changes to the DNS node settings can be a cumbersome process. Operators also have no flexibility in selecting SGW / PGW / GGSN nodes based on certain characteristics of the User Equipment (UE), such as location, mobile country code, etc.
[0050] For example, the MME uses a simple polling algorithm to load balance PDN-sessions across these SGW / PGW / SAEGW network elements. The MME network element queries the DNS network element to return a list of SGW / PGW / SAEGW network elements that it can select for a given APN, subscriber tracking area and other parameters, and then the MME network element polls for new PDN-sessions across the given list of SGW / PGW / SAEGW network elements. There are few problems with this approach:
[0051] 1. The MME assumes that all SGW / PGW / SAEGW network elements support the same PDN-session capacity and PDN-session signaling rate. However, this assumption may not be correct when the wireless operator selects multiple SGW / PGW / SAEGW network element suppliers, and also when the operator deploys different capacity and performance SGW / PGW / SAEGW network elements from the same supplier. Additionally, in a virtualized SGW / PGW / SAEGW environment, the capacity of the network elements changes dynamically when VMs or containers in the network element cluster are added or removed. Thus, the assumption of static PDN-session capacity and PDN-session signaling rate is invalid, especially in a virtualized EPC.
[0052] 2. In the wireless operator network, for hosting PDN-sessions on SGW / PGW / SAEGW network elements and since each MME does not know the exact number of PDN-sessions associated with each SGW / PGW / SAEGW network element, there may be multiple MMEs, and each MME selects the SGW / PGW / SAEGW network element independently. In this way, the MME does not have accurate current load information about each SGW / PGW / SAEGW network element for load balancing.
[0053] 3. If the SGW / PGW / SAEGW network element returns "system busy" for the request, the MME selects an alternative network element, which increases the latency of the session establishment.
[0054] 4. During planned SGW / PGW / SAEGW downtime (eg during software upgrades), the downed network element cannot inform the MME not to select it for new sessions.
[0055] Another problem that operators face when they work with multiple vendors to deploy SGW / PGW / GGSN in their networks is subscriber roaming. When subscribers roam outside the operator's network, they use the SGW present in the roaming network to establish a session to the PGW located in the operator's network. The subscriber's billing information is provided by the roaming partner. The operator does not have the ability to reconcile these billing records with the billing records generated by the nodes in the operator's network in real time to ensure the integrity of the billing data.
[0056] Figure 1 is a system diagram illustrating a networked system. Figure 1 A user equipment (UE) 102, an evolved Node B (eNodeB) 104, an evolved packet data gateway (ePDG) 108, a serving gateway (SGW) / serving GPRS support node (SGSN) 112, a packet data network gateway (PGW) / gateway GPRS support node (GGSN) 114, a mobility management entity (MME) 116, a traffic detection function (TDF) 118, a domain name system (DNS) 120, and a Gi network 130 are shown.
[0057] UE 102 is connected to the networking system 100 through eNodeB 104. UE 102 includes a computing device (e.g., mobile phone, tablet, laptop) configured to connect to a mobile data network. eNodeB 104 is the radio portion of a cell site. A single eNodeB 104 may include multiple radio transmitters, receivers, control sections, and power supplies. eNodeB 104 may be backhauled to MME 116 and SGW / SGSN 112. Backhaul is the process of delivering packets or communication signals to a separate location over a relatively long distance for processing. SGW / SGSN 112 routes and forwards user data packets, while also acting as a mobility anchor for the user plane during handover between eNodeBs. In some network implementations, only one of the SGW and SGSN is present in the network. For example, the SGSN acts as a service node in a general packet radio service (GPRS) network, while the SGW acts as a service node in an evolved packet core (EPC) network. In some network implementations, both the SGSN and the SGW exist as separate nodes, for example, where the SGW acts as a service node for traffic from the SGSN. The MME 116 is a control node in the networking system 100. The MME 116 handles paging procedures and marking procedures as well as retransmissions. The MME 116 may not exist in a GPRS network because the SGSN 112 includes functionality similar to that of the MME 116.
[0058] When UE 102 is attached to the network, multiple control messages are exchanged between network elements to create a data session (e.g., 3G session, 4G session) and provide data connectivity to UE 102. As explained above, eNodeB 104 can be backhauled to MME 116 and SGW / SGSN 112. SGW / SGSN 112 routes and forwards user packets to PGW / GGSN 114. In some network implementations, only one of PGW and GGSN exists in the network. For example, GGSN acts as a service node in a general packet radio service (GPRS) network, and PGW acts as a service node in an evolved packet core (EPC) network. PGW / GGSN114 routes packets to and from Gi network 116. Gi network 116 can include any enterprise server or origin server. Gi network 116 can be connected to the Internet or any other third-party server.
[0059] The ePDG 108 connects the UE 102 from an untrusted network (e.g., a Wi-Fi network) to a trusted network. The ePDG 108 typically routes data from the untrusted network to the PGW in the EPC network. The ePDG 108 may include processing capabilities, including deep packet inspection, policy enforcement, and tunnel authentication.
[0060] TDF 118 is a traffic management network element introduced with LTE Release 11. TDF 118 analyzes data session traffic and implements policies set by a provider (also referred to herein as a service provider or operator).
[0061] DNS120 converts domain names into IP addresses and can be queried by any of the network nodes described above. DNS120 can be implemented in a server connected to the network node via an internal or external network connection. Each of the network nodes described above can be configured to communicate with DNS120.
[0062] Figure 2 is a system diagram illustrating a networked system according to some embodiments of the present disclosure. Figure 2 A user equipment (UE) 102, an evolved Node B (eNodeB) 104, an evolved packet data gateway (ePDG) 204, a serving gateway (SGW) / serving GPRS support node (SGSN) 206, a packet data network gateway (PGW) / gateway GPRS support node (GGSN) 208, a service detection function (TDF) 209, a mobility management entity (MME) 202, a GPRS tunneling protocol agent (GTP agent) 210, a domain name system (DNS) 120 and a Gi network 130 are shown. Figure 2 Also shown are intelligent selection engines 212, 214, 216, 220 associated with MME 202, ePDG 204, SGW / SGSN 206 and PGW / GGSN 208, GTP proxy 210 and GTP proxy 211, respectively, and a separate session management engine 218.
[0063] Each of the intelligent selection engines 212, 214, 216, 218, 220 may be deployed independently or in conjunction with another intelligent selection engine.
[0064] In addition to having session management capabilities (which will be described in more detail below), MME 202, ePDG 204, SGW / SGSN 206, and PGW / GGSN 208 are similar to MME 116, ePDG 108, SGW / SGSN 112, and PGW / GGSN 114. In short, the intelligent selection engine 212 integrated with MME 202 enables MME 202 to select a SGW among multiple SGWs based on SGW availability and load; the intelligent selection engine 214 integrated with ePDG 204 enables ePDG 206 to select a SGW among multiple SGWs; 204 is capable of selecting a PGW among multiple PGWs based on PGW availability and load; an intelligent selection engine 216 integrated with a control plane for SGW, PGW or TDF (also referred to as SGW-C, PGW-C or TDF-C) enables SGW-C, PGW-C or TDF-C to select a user plane among multiple user plane SGWs, PGWs or TDFs (also referred to as SGW-U, PGW-U or TDF-U) based on user plane availability and load, respectively; an intelligent selection engine 218 as an independent node can override the MME's selection of SGW based on SGW availability and load; and an intelligent selection engine 220 integrated with a GTP agent 210 enables the GTP agent 210 to select a PGW and / or GGSN among multiple PGWs and / or GGSNs based on PGW / GGSN availability and load.
[0065] When there is a mismatch in the protocol (e.g., different APNs), such as when the SGW / SSGN and PGW / GGSN are under the control of different mobile network operators (MNOs), the GTP proxy 210 converts the data routed from the SGW / SSGN to the PGW / GGSN. As described in more detail below, in addition to load balancing, the GTP proxy 210 can also be used to hide the network topology, convert session information between different LTE version formats, map APNs, and provide enhanced charging capabilities. Although the GTP proxy 210 is in Figure 2 Although shown as having only the intelligent selection engine 220, the GTP agent 210 may also include other modules.
[0066] The GTP agent 211 converts data routed from the MME 202 to the SGW / SSGN 206. As described in more detail below, the GTP agents 210, 211 can be inserted between the SGW / SSGN and the PGW / GGSN or between the MME and the SGW to hide the network topology, convert session information between different LTE version formats, map the APN, and provide enhanced charging capabilities and load balancing. The GTP agents 210, 211 act as GTP endpoints supporting the Gn interface, Gp interface, S5 interface, S8 interface, and S11 interface between the SGW / SGSN and the PGW / GGSN and between the MME and the SGW. The GTP agents 210, 211 support the termination and relaying of GTP-C services and GTP-U services. When inserted between nodes, the GTP agents 210, 211 hide the network topology by selecting the SGW and PGW nodes to forward the session establishment request. Similarly, when inserted between the SGW / SGSN and the PGW / GGSN, the GTP proxy 210, 211 also selects the PGW / GGSN node to forward the session establishment request. Fig.21 As described in more detail, the intelligent selection engine can also be implemented in a 5G network or other network architectures.
[0067] Figure 3 is a system diagram showing an MME including an intelligent selection engine according to some embodiments of the present disclosure. Figure 3 The MME 202 is shown including the intelligent selection engine 212 , the DNS 120 , the SGW 112 , the PGW 114 , a DNS 120 to MME 202 interface 302 , an MME 202 to SGW 112 interface 304 , and a SGW 112 to PGW 114 interface 306 .
[0068] The intelligent selection engine 212 runs as a microservice or library in the MME 202. The intelligent selection engine 212 can run as a container or virtual machine within the MME 202 or on a blade. The MME 202 interfaces with the DNS 120 using a DNS query 302. As described in more detail below, the MME 202 queries the DNS 120 to select a SGW 112 for a given UE 102 session based on the availability and load of the SGW. The MME 202 interfaces with the SGW 304 via S11 and GTP-v2, and the SGW interfaces with the PGW 306 via S5 / S8 and GTP-v2. As described in more detail below, the intelligent selection engine 212 allows the MME 202 to select the SGW 112 and PGW 114 based on criteria such as priority, load, and peer status.
[0069] Figure 4 is a system diagram showing an ePDG including an intelligent selection engine according to some embodiments of the present disclosure. Figure 4 The ePDG 204 is shown including the intelligent selection engine 212 , the DNS 120 , the PGW 114 , the DNS 120 to ePDG interface 402 , and the ePDG 204 to PGW 114 interface 404 .
[0070] The intelligent selection engine 214 runs as a microservice or library in the ePDG 204. The intelligent selection engine 214 can run as a container or virtual machine within the ePDG 204 or on a blade. The ePDG 204 interfaces with the DNS 120 using a DNS query 402. As described in more detail below, the ePDG 204 queries the DNS 120 to select a PGW 114 for a given UE 102 session based on the availability and load of the PGW. The ePDG 204 interfaces with the PGW 114 via 52b and GTP-v2.
[0071] Figure 5 is a system diagram illustrating a control plane for a SGW, PGW or TDF including an intelligent selection engine according to some embodiments of the present disclosure. Figure 5 A control plane 502 for SGW-C / PGW-C / TDF-C including an intelligent selection engine 216, DNS120, a user plane SGW-U / PGW-U / TDF-U 504, a DNS120 to control plane SGW-C / PGW-C / TDF-C interface 512, and a control plane SGW-C / PGW-C / TDF-C to user plane SGW-U / PGW-U / TDF-U interface 514 are shown.
[0072] The intelligent selection engine 216 runs as a microservice or library in one of the SGW-C, PGW-C, or TDF-C 502. The intelligent selection engine 216 can run as a container or virtual machine within the control plane 502 of the SGW-C, PGW-C, or TDF-C or on a blade. The control plane 502 of the SGW-C, PGW-C, or TDF-C uses DNS queries 512 to interface with the DNS 120. As described in more detail below, the control plane 502 of the SGW-C, PGW-C, or TDF-C queries the DNS 120 to select the user plane SGW-U, PGW-U, or TDF-U 504 for a given UE 102 session based on the availability and load of the user plane SGW-U, PGW-U, or TDF-U. The control plane 502 of the SGW-C, PGW-C or TDF-C interfaces with the user plane SGW-U, PGW-U or TDF-U 504 through the Packet Forwarding Control Plane Protocol (PFCP) and Sxa, Sxb or Sxc.
[0073] Figure 6 is a system diagram illustrating an independent intelligent selection engine according to some embodiments of the present disclosure. Figure 6 The intelligent selection engine 218, DNS120, SGW 112, PGW 114, MME 116, DNS120 to MME 116 interface 602, MME116 to intelligent selection engine 218 interface 602, DNS120 to intelligent selection engine 218 interface, intelligent selection engine 218 to SGW112 interface 608, and SGW 112 to PGW 114 interface 610 are shown.
[0074] The intelligent selection engine 218 runs as a microservice or on a standalone node (e.g., a virtualized node or a rack-based node on a cloud). The intelligent selection engine 218 interfaces with the DNS 120 using a DNS query 606 and interfaces with the MME 116 using S11 / Access and GTP-v2. As described in more detail below, the MME 116 queries the DNS 120 to select the SGW 112 for a given UE 102 session, the intelligent selection engine 218 sniffs the CreateSessionRequest GTP-C packet from the MME 116, and overwrites the selected SGW / PGW IP address of the MME in the CreateSessionRequest message based on the availability and load of the SGW and PGW. The intelligent selection engine 218 interfaces with the SGW 112 via S11 / Core and GTP v-2, and the SGW 112 interfaces with the PGW 114 via S5 / S8 and GTP-v2.
[0075] Figure 7 is a system diagram showing a GTP agent including an intelligent selection engine according to some embodiments of the present disclosure. Figure 7 The GTP agent 210 is shown, which includes an intelligent selection engine 210, DNS120, Internet Protocol (IP) 720, SGSN 732, SGW 734, GGSN 736, PGW 738, DNS120 to GTP agent 210 interface 402, GTP agent 210 to IP 720 access interface 704, GTP agent 210 to IP 720 core interface 710, SGSN 732 to IP 720 interface, SGW734 to IP 720 interface, GGSN 736 to IP 720 interface and PGW 738 to IP 720 interface 714.
[0076] In addition to SGSN 732, SGW 734, PGW 738 and PGSN 706 Figure 7 734 are similar to SGW / SGSN 112, and GGSN 736 and PGW 738 are similar to PGW / PGSN 114, except that they are shown as separate nodes in FIG.
[0077] The GTP proxy 210 is inserted between the SSGN 702 / SGW 734 and the GGSN 736 / PGW 738. The GTP proxy 201 includes an intelligent selection engine 210 that runs as a microservice or library in the GTP proxy 210. The intelligent selection engine 210 can run as a container or virtual machine within the GTP proxy 210 or on a blade. The GTP proxy 210 interfaces with the DNS 120 using a DNS query 702. As described in more detail below, the ePDG 204 queries the DNS 120 to select the PGW 114 or GGSN 736 for a given UE 102 session based on the availability and load of the PGW and GGSN and the received SGSN information and SGW information. Based on the query, the DNS 120 can provide a list of nodes that can support the query. SGSN 732 interfaces with IP 720 via Gn / Gp 706, SGW 734 interfaces with IP 720 via S5 / S8, GTP proxy 210 interfaces with IP 720 via Gn / Gp / S5 / S8-C and GTP-CV0 / V1 / v2 or Gn / Gp / S5 / S8-U and GTP-UV1 on access interface side 704 and on core interface side 706, IP 720 interfaces with GGSN via Gn / Gp and interfaces with PGW 738 via S5 / S8. Solid lines refer to control plane interfaces, while dashed lines refer to user plane interfaces.
[0078] Figure 8 is a schematic diagram illustrating a peer selector according to some embodiments of the present disclosure. Figure 8 Smart selection engines 212, 214, 216, 218, 220, subscriber analyzer 802, peer selector 804, priorities 812, keys 814, peer FQDNs (fully qualified domain names) 816, peer lists 818, and peer selection algorithm 820 are shown.
[0079] The intelligent selection engines 212, 214, 216, 218, 220 include similar modules and logic for selecting peer nodes, respectively, and will be discussed collectively as the intelligent selection engine 800. The intelligent selection engine 800 includes a peer selector 804. The peer selector can be one of many modules that can be used in the intelligent selection engine 800 or a subscriber analyzer 802. As described in more detail below, the peer selector 804 can be selected by the intelligent selection engine 800. The peer selector 804 then selects a peer node based on the availability and load of the peer node. The intelligent selection engine 800 can be associated with one or more peer selectors 804.
[0080] The peer selector 804 may be supplied by an operator based on a use case or workflow. Each peer selector 804 may have one or more keys 814. In some embodiments, each key 814 may be represented as a key pair (e.g., {key type, key value}). The key 814 may identify the type of service (e.g., LTE version type, location, GTP agent name). For example, for a GTP agent, the key may include an apn name (apn-name), a gtp agent name (gtp-proxy-name), an eutran-cell-global-id, a home-plmnid, an imei-range, an imsi-range, a rat type (rat-type), a routing area id (routing-area-id), a service area id (service-area-id), a service nw-plmnid (serving-nw-plmnid), a tracking area id (tracking-area-id), a ue-ip address (ue-ip-address), and a service nw-mcc (serving-nw-mcc). The keys may be combined using an OR operation and an AND operation. For example, a peer selector may be selected based on a key that matches both the LTE version type and the location. Priority 812 indicates the order in which peer selectors 804 are selected over other peer selectors. For example, if two peer selectors have matching keys associated with a service, the peer selector with the higher priority is selected.
[0081] Once a peer selector is selected, a peer node is selected from one of the peer FQDNs 816 and the peer list 818. As described in more detail below, if the peer selector 804 maps to a peer FQDN, the intelligent selection engine 800 sends a DNS query to obtain a peer list associated with the peer FQDN. If the peer selector is not mapped to a peer FQDN or if a DNS server is not configured, the peer selector can use a predefined peer list associated with the peer selector. Once a peer list is selected, a peer node from the peer list is selected using a peer selection algorithm 820, which takes into account both peer availability and load, as described in more detail below.
[0082] Fig. 9 is a flow chart illustrating a process of selecting a peer list according to some embodiments of the present disclosure.
[0083] Referring to step 902, the intelligent selection engine 800 receives a trigger for selecting a peer node for a subscriber session. The trigger may include an indication of session traffic traversing nodes associated with the intelligent selection engine 800, or a specific request to select a peer node associated with the session traffic or provided by an operator.
[0084] Referring to step 904, the intelligent selection engine 800 selects a peer selector 804. In some embodiments, the peer selector 804 with the highest priority and a matching subscriber key is selected. As part of selecting a peer selector, the intelligent selection engine 800 may analyze subscriber session data to determine parameters that match its key type (e.g., the location of the subscriber). In some embodiments, the parameters in the subscriber session data are determined by another node and provided to the intelligent selection engine 800. Once one or more peer selectors are selected based on the peer selector having a key type that matches the user subscriber parameter, a peer selector is selected from the one or more peer selectors based on the peer selector having a high priority. Examples of peer selectors including matching keys with different priority levels include: if the primary peer node is reachable, the primary peer node is selected; otherwise, the secondary peer node is selected. The primary peer node is associated with a higher priority peer selector, while the secondary peer node is associated with a low priority peer selector.
[0085] Referring to step 906, the intelligent selection engine 800 determines whether the selected peer selector is associated with the FQDN (906). In some embodiments, the association can be configured by the operator. If the selected peer selector is associated with the FQDN, the intelligent selection engine 800 determines whether the DNS server is configured (908). In some embodiments, the DNS server is configured when the primary DNS service or the secondary DNS service is supplied and reachable. If the DNS server is configured, the intelligent selection engine 800 requests the DNS server to return a peer list associated with the FQDN (910). If the DNS server is not configured, if the peer selector is not associated with the FQDN, the intelligent selection engine 800 uses a peer list from the peer selector (912). As mentioned above, the peer list from the peer selector (912) can be pre-configured by the operator or dynamically associated in other ways.
[0086] Referring to step 914, once the peer list is selected, the intelligent selection engine 800 selects a peer node from the peer list. Fig.10 and Fig.10 As described in more detail in the accompanying text, the intelligent selection engine 800 selects peer nodes using a peer selection algorithm that considers both peer availability and load.
[0087] Fig.10is a flow chart illustrating a process of selecting a peer node from a peer list according to some embodiments of the present disclosure.
[0088] Referring to step 1002, the intelligent selection engine 800 receives a peer list (e.g., as described above in Fig. 9 peer list selected as described in .
[0089] Referring to step 1004, the intelligent selection engine 800 excludes unreachable peer nodes from the peer list. A peer node may be considered unreachable if it has failed, is intentionally shut down for maintenance or updating, or has lost connection with the network for some other reason. The intelligent selection engine 800 determines which peer nodes are unreachable by receiving messages from other nodes regarding the status of the peer nodes. Other nodes may determine the status of a peer node by sending a heartbeat message to the peer node to test whether the link is still active.
[0090] Referring to step 1006, the intelligent selection engine 800 excludes overloaded or about-to-shutdown peer nodes from the peer list. If the peer node has reached or is reaching maximum capacity or is experiencing software corruption issues, the peer node can be considered overloaded. The intelligent selection engine 800 determines which peer nodes are overloaded or about to shut down in a similar manner to its detection of whether the peer node is reachable. For example, the intelligent selection engine 800 can determine which peer nodes are overloaded by receiving messages about the status of the peer nodes from other nodes. Other nodes can determine the status of the peer nodes by sending heartbeat messages to the peer nodes to test whether the peer nodes are overloaded. Peer nodes can also be excluded based on management instructions. For example, if a node is being repaired or upgraded, the node can be removed and is considered to be the intelligent selection engine 800 (because it is unavailable).
[0091] Referring to step 1008, the intelligent selection engine 800 groups the remaining peer nodes based on the load level thresholds of the remaining peer nodes. For example, peer nodes can be associated with different loads (e.g., in terms of actual load, total capacity percentage, or load level) and associated load levels (e.g., low load level, medium load level, or high load level). The load level can be determined from the load. For example, peer nodes with a high total capacity percentage can be associated with a high load level. The load level threshold can be preset or determined based on a combination of the load and load level associated with each node. For example, peer nodes associated with a low level can have a load level threshold of 30%; peer nodes associated with a medium load level can have a load level threshold of 60%; and peer nodes associated with a high load level can have a load level threshold of 90%. The percentage can be preset or can be determined based on peer clusters observed in the system. For example, the load level can be determined based on the average load associated with the peer node or the median load associated with peer nodes in different groups, or other similar methods. Multiple groups or deviations close to a threshold value may be determined for the purpose of grouping. Other grouping techniques (e.g., clustering, Bayesian classification) may also be used. In some embodiments, one result of grouping is to establish peer groups from least loaded to most loaded.
[0092] Referring to step 1010, the intelligent selection engine 800 selects a peer node from the least loaded peer group. The process for selecting a peer node from the least loaded peer group may include round robin, weighted round robin, or hashing.
[0093] In addition to load balancing capabilities, the GTP agent with intelligent selection engine 220 can also be used to hide network topology, convert session information between different LTE version formats, map APNs, and provide enhanced charging capabilities. GTP agent 210 is independent of hardware and management programs, and can be integrated with hardware and management programs (such as HPC7000, DL360 / 380, IBM Pureflex, ATCA, VMWare, KVM, and Openstack). In some embodiments, the functions of GTP agent 210 can be virtualized, and virtual machines can be added to the GTP agent cluster without affecting other network elements. GTP agent 210 may include three different types of virtual machines: management control module (MCM), control plane module (CPM), and subscriber session module (SSM). MCM is responsible for the general operation and management of GTP agent, including operation, administration and management (OA&M), command line interface (CLI), and network configuration protocol (NETCONF). CPM is responsible for dynamic routing, session management, GTP-C, call control, OFCS client, and OCS client. SSM is responsible for logical IP interface termination, bearer GTP-U tunnel termination, and charging. In some embodiments, the intelligent selection engine 220 may run on a CPM.
[0094] Fig.11 1 is a system diagram showing a networked system including a GTP proxy with an intelligent selection engine according to some embodiments of the present disclosure. The networked system 1100 shows a GTP proxy 210, which includes an intelligent selection engine 220, an Internet Protocol (IP) 720, an SGSN 732, an SGW 734, a GGSN 736, a PGW 738, an enhanced messaging service (EMS) 1110, an online charging system (OCS) 1112, an offline charging system (OFCS) 1114, a GTP proxy 210 to IP 720 access interface 704, a GTP proxy 210 to IP 720 core interface 710, an SGSN 732 to IP 720 interface 706, an SGW 734 to IP 720 interface 708, a GGSN 736 to IP 720 interface 712, a PGW 738 to IP 720 interface 714, a GTP proxy 210 to IP 720 interface 1102 , GTP proxy 210 to EMS 1110 interface 1104 , IP 720 to OCS 1112 interface 1106 , and IP 720 to OFCS 1114 interface 1108 .
[0095] EMS 1110 interfaces with GTP agent 210 via NETCONF interface 1104. EMS 1110 is a messaging service that is an extension of the Short Message Service (SMS). EMS 1110 is similar to Multimedia Messaging Service (MMS) in that it supports messages containing both text and multimedia messages (e.g., audio, video, images). In contrast to MMS, EMS messages can be received by devices that are configured to receive only SMS messages. Although EMS 1110 is shown in the drawings, the GTP agent can also interface with MMS and SMS.
[0096] IP 720 interfaces with GTP agent 210 via IP Internet Protocol 1102, which connects GTP agent 210 with OCS 1112 and OFCS 1114. IP 720 is a router placed between GTP agent 210 and other network nodes. In some embodiments, GTP agent 210 is directly connected to other network nodes, and there is no router between GTP agent 210 and other network nodes. OCS 1112 and OFCS 1114 interface with IP 720 via Gy 1106 and Gz / Ga 1108, respectively. OCS 1112 allows the operator to charge subscribers in real time for data usage. OFCS 1114 generates billing records based on the subscriber's data usage. Both OCS 1112 and OFCS 1114 allow event-based charging and session-based charging.
[0097] As described in more detail below, having the GTP proxy 210 with the smart session engine 220 located between the SGSN 732 / SGW 734 and the GGSN 736 / PGW 738 may be advantageous due to topology hiding and flexibility in selecting a gateway node. For example, by moving the gateway node selection process from the MME to the GTP proxy 210, a different gateway node may be allowed to be selected without reconfiguring the DNS.
[0098] Fig.12is a system diagram illustrating a networking system including a GTP proxy with an intelligent selection engine according to some embodiments of the present disclosure. The networking system 1200 shows a GTP agent 210, which includes an intelligent selection engine 220, an Internet Protocol (IP) 720, an SGSN 732, an SGW 734, an enhanced messaging service (EMS) 1110, an offline charging system (OFCS) 1114, a GTP agent 210 to IP 720 access interface 704, a GTP agent 210 to IP 720 core interface 710, an SGSN 732 to IP 720 interface 706, an SGW 734 to IP 720 interface 708, an eNodeB to IP 720 interface 1202, an MME 116 to IP 720 interface 1204, a GTP agent 210 to IP 720 interface 1102, a GTP agent 210 to EMS 1110 interface 1104, and an IP 720 to OFCS 1114 interface 1108.
[0099] The eNodeB 104 interfaces with the IP 720 via the IP Internet Protocol 1202. The MME interfaces with the IP 720 via the IP Internet Protocol 1204.
[0100] As described in more detail below, Fig.11 and Fig.12 The systems shown together therein (and in particular, the location and capabilities of the GTP proxy) allow at least the following: 1) topology hiding and flexibility to add, delete or move network elements to different locations in the network without having to change DNS settings; 2) session state-aware processing of call data records (CDRs) generated to support coordination with SGW / SGSN / PGW / GGSN billing records; 3) enhanced SGW / PGW / GGSN (peer) selectors based on peer reachability, peer load, peer overload and UE session attributes (radio access technology (RAT) type, APN name, location and mobile country code (MCC) / mobile network code (MNC)); 4) APN manipulation, such as grouping multiple UE APNs into a single APN name; 5) interworking of different network elements with different standard versions by adding / suppressing GTP information elements (IEs); 6) allowing / denying selected GTP procedures (e.g., GTP V0, dedicated bearer); 7) flexibility in adding / removing GTP peer nodes (SGW / PGW); and 8) enabling workflow services (e.g., Deep Packet Inspection (DPI), Quality of Service (QoS) enforcement).
[0101] Fig.131 is a schematic diagram showing a GTP proxy positioned between an SGW and a PGW to implement a create session request according to some embodiments of the present disclosure. In this exemplary embodiment, the IP of SGW 1334 is 10.11.12.15, the IP address of GTP proxy 1310 is 10.10.10.10, and the IP address of PGW 1338 is 10.10.12.13. In addition to having assigned IP addresses, SGW 1334, GTP proxy 1310, and PGW 1338 have the same functions and structures as SGW 734, GTP proxy 210, and PGW 738, respectively. Fig.13 Also shown are create session request 1302, create session request 1304, create session response 1312, and create session response 1314.
[0102] The create session request 1312 transmitted from the SGW 1334 to the GTP proxy 1310 includes the following information:
[0103] Create Session Request
[0104] GTP header TEID 0
[0105] Destination PGW IP 10.10.10.10
[0106] Sender FTEID IP 10.11.12.15
[0107] Sender FTEID TEID 500
[0108] The GTP header TEID refers to the tunnel endpoint identifier associated with the Create Session Request message. The GTP header TEID identifies the address of the recipient of the communication during the session. Each session and recipient can be associated with a different GTP header TEID. In some embodiments, the GTP header TEID can be reused after a period of time (e.g., expiration of a timer). When the recipient TEID is unknown when the session is established, the value of the GTP header TEID is usually "0".
[0109] The destination PGW IP refers to the destination PGW IP address. As shown above, the destination PGW IP address is 10.10.10.10, which is the IP address of GTP proxy 1310, not the IP address of PGW 1338, which is 10.10.12.13. As described in more detail below, having the GTP proxy intercept the create session request from SGW 1334 allows the GTP proxy to further analyze, convert, and direct the session request.
[0110] The sender FTEID IP refers to the fully qualified endpoint identifier IP address of the sender. This is the IP address associated with the transport node, which in this case is the SGW 1334 with an IP address of 10.11.12.15.
[0111] The sender FTEID TEID refers to the fully qualified endpoint identifier Tunnel Endpoint Identifier of the sender. The sender FTEID TEID is similar to the GTP Header TEID except that the sender FTEID TEID identifies a specific session associated with the sender rather than a session associated with the receiver.
[0112] After receiving Create Session Request 1302, GTP Proxy 1310 transmits the following Create Session Request 1304 to PGW 1338:
[0113] Create Session Request
[0114] GTP header TEID 0
[0115] Destination PGW IP 10.10.12.13
[0116] Sender FTEID IP 10.10.10.10
[0117] Sender FTEID TEID 530
[0118] The GTP header TEID is 0. As mentioned above, the GTP header TEID is 0 because the recipient TEID is unknown (because this is a create session request).
[0119] The destination PGW IP is 10.10.12.13, which is the IP address of PGW 1338. The sender FTEID IP is 10.10.10.10, which is the address of GTP proxy 1310.
[0120] After receiving Create Session Request 1304, PGW 1338 transmits the following Create Session Response 1312 to GTP Proxy 1310:
[0121] Create Session Response
[0122] GTP header TEID 530
[0123] Destination SGW IP 10.10.10.10
[0124] Sender FTEID IP 10.10.12.13
[0125] Sender FTEID TEID 910
[0126] The GTP header TEID is 530, which is an ID associated with the GTP proxy.
[0127] Destination SGW IP refers to the destination SGW IP address. As shown above, the destination SGW IP address is 10.10.10.10, which is the IP address of GTP proxy 1310, not the IP address of SGW 1334, which is 10.11.12.15. As described in more detail below, having GTP proxy 1310 intercept the create session response from PGW 1338 allows the GTP proxy to further analyze, transform, and direct the session response.
[0128] The sender FTEID IP is 10.10.12.13, which is the IP address of PGW 1338. The sender FTEID TEID is 910, which is the tunnel ID of PGW.
[0129] After receiving Create Session Response 1312, GTP Proxy 1310 transmits the following Create Session Response 1314 to SGW 1334:
[0130] Create Session Response
[0131] GTP header TEID 500
[0132] Destination SGW IP 10.11.12.15
[0133] Sender FTEID IP 10.10.10.10
[0134] Sender FTEID TEID 550
[0135] The GTP header TEID is 500, which is different from the GTP header TEID in the create session response 1312.
[0136] The destination SGW IP is 10.11.12.15, which is the IP address of SGW 1334. The sender FTEID IP is 10.10.10.10, which is the address of GTP agent 1310. The sender FTEID TEID of 550 is the tunnel ID of GTP agent 1310.
[0137] Fig.1414 is a schematic diagram showing a GTP proxy positioned between two SGWs and a PGW in order to implement an internal 4G handover with SGW changes according to some embodiments of the present disclosure. In this exemplary embodiment, the IP of SGW1 1402 is 10.11.12.15, the IP of SGW2 1404 is 10.11.12.14, the IP address of GTP proxy 1410 is 10.10.10.10, and the IP address of PGW 1406 is 10.10.12.13. In addition to having assigned IP addresses, SGW1 1402 / SGW 2 1404, GTP proxy 1410, and PGW 1406 have the same functions and structures as SGW 734, GTP proxy 210, and PGW 738, respectively. Fig.14 Also shown are create session request 1412, create session request 1414, create session response 1416, create session response 1418, modify bearer request 1420, modify bearer request 1422, modify bearer response 1424, and modify bearer response 1426.
[0138] Create session request 1412, create session request 1414, create session response 1416, and create session response 1418 similarly have the same content and functionality as create session request 1302, create session request 1304, create session response 1312, and create session response 1314. The description of create session request 1302, create session request 1304, create session response 1312, and create session response 1314 is applicable to create session request 1412, create session request 1414, create session response 1416, and create session response 1418.
[0139] The modifier bearer request 1420 transmitted from SGW2 1404 to the GTP proxy 1410 includes the following information:
[0140] Modify Bearer Request
[0141] GTP header TEID 550
[0142] Sender FTEID IP 10.11.12.14
[0143] Sender FTEID TEID 600
[0144] Destination PGW IP 10.10.10.10
[0145] As mentioned above, the GTP header TEID refers to the tunnel endpoint identifier associated with the modify bearer request message.
[0146] The destination PGW IP refers to the destination PGW IP address. As shown above, the destination PGW IP address is 10.10.10.10, which is the IP address of the GTP proxy 1410, not the IP address of PGW1 1406, which has an IP address of 10.10.12.13. As described in more detail below, having the GTP proxy intercept the modify bearer request from the SGW 1404 allows the GTP proxy to further analyze, convert, and direct the session request.
[0147] The sender FTEID IP refers to the fully qualified endpoint identifier IP address of the sender. This is the IP address associated with the transport node, which in this case is the SGW 1404 whose IP address is 10.11.12.14.
[0148] Sender FTEID TEID refers to the fully qualified endpoint identifier of the sender Tunnel Endpoint Identifier. Sender FTEID TEID is 600, which is the tunnel ID of the GTP proxy.
[0149] After receiving the modify bearer request 1420, GTP proxy 1410 transmits the following modify bearer request 1422 to PGW1 1406:
[0150] Modify Bearer Request
[0151] (New PLMN ID / ULI / RAT / REID)
[0152] GTP header TEID 910
[0153] Sender FTEID TEID 540
[0154] The GTP header TEID is 910 and the sender FTEID TEID is 540.
[0155] The GTP agent 1410 may specify a new PLMID, ULI, RAT, or REID.
[0156] After receiving the Modify Bearer Request 1422, PGW 1406 transmits the following Modify Bearer Response 1424 to GTP Proxy 1410:
[0157] Modify the bearer response
[0158] GTP header TEID 540
[0159] Destination SGW IP 10.10.10.10
[0160] The GTP header TEID is 540.
[0161] As described above, the destination SGW IP refers to the destination SGW IP address. As shown above, the destination SGW IP address is 10.10.10.10, which is the IP address of the GTP proxy 1410, not the IP address of the SGW 1404, which is 10.11.12.14. As described in more detail below, having the GTP proxy 1410 intercept the modified bearer response from the PGW 1406 allows the GTP proxy to further analyze, transform, and direct the session response.
[0162] After receiving the modify bearer response 1424, the GTP proxy 1410 transmits the following modify bearer response 1426 to SGW2 1404:
[0163] Modify the bearer response
[0164] GTP header TEID 600
[0165] Destination SGW IP 10.11.12.14
[0166] The GTP header TEID is 600, which is different from the GTP header TEID in the modify bearer response 1424.
[0167] The destination SGW IP is 10.11.12.14, which is the IP address of SGW2 1404. The sender FTEID IP is 10.10.10.10, which is the address of GTP proxy 1410.
[0168] Fig.15 1 is a schematic diagram showing a GTP proxy positioned between two SGWs and a PGW to implement 4G to 3G handover according to some embodiments of the present disclosure. In this exemplary embodiment, the IP of SGW 1502 is 10.11.12.15, the IP of SGSN 1504 is 10.11.12.14, the IP addresses of GTP proxy 1510 are 10.10.10.10 and 10.10.10.11, and the IP address of PGW 1506 is 10.10.12.13. In addition to having assigned IP addresses, SGW 1502, SGSN 1504, GTP proxy 1510, PGW 1506 have the same functions and structures as SGW 734, SGSN 732, GTP proxy 210, and PGW 738, respectively. Fig.15 Also shown are create session request 1512, create session request 1514, create session response 1516, create session response 1518, update PDP context request 1520, update PDP context request 1522, update PDP context request 1524, and update PDP context response 1526.
[0169] Create session request 1512, create session request 1514, create session response 1516, and create session response 1518 similarly have the same content and functionality as create session request 1302, create session request 1304, create session response 1312, and create session response 1314. The description of create session request 1302, create session request 1304, create session response 1312, and create session response 1314 is applicable to create session request 1512, create session request 1514, create session response 1516, and create session response 1518.
[0170] The updated PDP context 1520 transmitted from the SGSN 1504 to the GTP proxy 1510 includes the following information:
[0171] Update PDP Competition Request
[0172] GTP header TEID 550
[0173] Sender FTEID IP 10.11.12.14
[0174] Sender FTEID TEID 600
[0175] Destination GGSN IP 10.10.10.10
[0176] As mentioned above, the GTP header TEID refers to the tunnel endpoint identifier associated with the Update PDP Context Request message. The value of the GTP header TEID is 550.
[0177] Destination GGSN IP refers to the destination GGSN IP address. As shown above, the destination GGSN IP address is 10.10.10.10, which is the IP address of GTP proxy 1510. As described in more detail below, having the GTP proxy intercept the Update PDP Context Request from SGSN 1504 allows GTP proxy 1510 to convert requests originally intended for the GGSN node to PGW 1506.
[0178] The sender FTEID IP refers to the fully qualified endpoint identifier IP address of the sender. This is the IP address associated with the transport node, which in this case is the SGSN 1504 whose IP address is 10.11.12.14.
[0179] Sender FTEID TEID refers to the fully qualified endpoint identifier of the sender tunnel endpoint identifier. The sender FTEID TEID is 600.
[0180] After receiving Update PDP Context Request 1520, GTP Proxy 1510 transmits the following Update PDP Context Request 1522 to PGW 1506:
[0181] Update PDP context request
[0182] (New PLMNID / ULI / RAT / TEID)
[0183] GTP header TEID 910
[0184] Sender FTEID IP 10.10.10.11
[0185] Sender FTEID TEID 540
[0186] The GTP header TEID is 910, which is different from the GTP header TEID in the Update PDP Context Request 1520. The sender FTEID TEID is 540.
[0187] The GTP agent 1510 may specify a new PLMID, ULI, RAT, or TEID.
[0188] After receiving Update PDP Context Request 1522, PGW 1506 transmits the following Update PDP Context Response 1524 to GTP Proxy 1510:
[0189] Update PDP Context Response
[0190] GTP header TEID 540
[0191] Destination SGSN IP 10.10.10.11
[0192] The GTP header TEID is 540.
[0193] As described above, the destination SGW IP refers to the destination SGSN IP address. As shown above, the destination SGSN IP address is 10.10.10.10, which is the IP address of the GTP proxy 1510, rather than the IP address of the SGSN 1504, which is 10.11.12.14. As described in more detail below, having the GTP proxy 1410 intercept the Update PDP Context Response from the PGW 1406 allows the GTP proxy to convert the response from the PGW 1506 into a response that can be received by the SGSN 1504.
[0194] After receiving Update PDP Context Response 1524, GTP Proxy 1510 transmits the following Update PDP Context Response 1526 to SGSN 1504:
[0195] Update PDP Context Response
[0196] GTP header TEID 600
[0197] Destination SGSN IP 10.11.12.14
[0198] The GTP header TEID is 600, which is different from the GTP header TEID in the Update PDP Context Response 1524.
[0199] The destination SGSN IP is 10.11.12.14, which is the IP address of SGSN 1504.
[0200] Fig.16 is a table showing parameters maintained by a GTP agent for switching tunnels when processing GTP-U traffic and GTP-C traffic according to some embodiments of the present disclosure. Fig.16 Packet type 1602, network element type 1604, network element TEID 160, GTP proxy TEID 1608, network element restart counter 1610, and GTP proxy restart counter 1612 are shown.
[0201] Packet type 1602 refers to whether the packet is GTP-U (data traffic) or GTP-C (eg, signaling message).
[0202] GTP Proxy TEID 1608 refers to the tunnel endpoint identifier for the GTP Proxy. As described in more detail below, each unique combination of packet type, network element type, and TEID is associated with a unique GTP Proxy TEID.
[0203] The GTP entity (e.g., PGW or SGW or GTP proxy) maintains two restart counters. One restart counter is the network element restart counter 1610, which is the counter sent by the peer node, and the other restart counter is the GTP proxy restart 1612, which is the GTP proxy's own counter. When an entity restarts, it immediately increments its restart counter and sends the new value in a subsequent echo request or response.
[0204] The network element type 1604 and the network element TEID 1606 associate a TEID with the network element.
[0205] For example, when a GTP-C packet is received from SGW1 and forwarded to PGW1, the GTP agent will set the TEID value in the header to 506.
[0206] For example, when a GTP-C packet received from PGW1 is forwarded to SGW1, the GTP agent sets the TEID value in the header to 502.
[0207] The GTP agent also maintains the restart counter value of the peer node for the GTP-C path in a table. This value refers to the restart counter maintained by the GTP-C entity in non-volatile memory, which is incremented whenever the node is restarted. The restart counter can be exchanged in the echo request and response messages. When the GTP agent detects that the restart counter value received from the peer node has changed, it can treat the peer node as a failed node and take action to clean up the session accordingly.
[0208] The intelligent selection engine 220 includes a workflow analyzer 1604. The workflow analyzer 1604 can be one of many modules or subscriber analyzers 802 that can be used with the intelligent selection engine 220. For example, and as described above, another type of subscriber analyzer 802 that can be used with the intelligent selection engine 220 is a peer selector 804. As described in more detail below, the workflow analyzer 1604 can be selected by the intelligent selection engine 220. The workflow analyzer 1604 then selects a workflow based on a matching key 1614 and a priority 1612. The intelligent selection engine 1604 can be associated with one or more workflow analyzers 1804.
[0209] Fig.17 is a flow chart illustrating processing of a create session request according to some embodiments of the present disclosure.
[0210] Referring to step 1712, after receiving the create session request, the GTP proxy 210 transmits it to the intelligent selection engine 220 that selects the PGW. In some embodiments, the create session request is directly routed to the peer selector 804. The create session request may be received from the SGW or the MME.
[0211] In some embodiments, the create session request is first routed to the workflow analyzer 1604 and then routed to the peer selector 804. By way of example, the create session request 1712 may include the following information:
[0212] Create Session Request
[0213] IMSI:12345678912345
[0214] Service network MCC = 310
[0215] APN Name = Internet 1 (internet1)
[0216] ISMI refers to International Mobile Subscriber Identity, which is a unique number associated with a subscriber. Serving Network MCC refers to Serving Network Mobile Country Code. Generally speaking, each country has a mobile country code. Another field that can be specified is the Mobile Network Code (MNC), which refers to the carrier network within a country. Typically, each MCC is associated with multiple MNCs.
[0217] The APN name refers to the access point name associated with the create session request. The APN identifies the gateway between the mobile network and another network (e.g., the Internet). As shown above, the APN name used for the create session request 1712 is Internet 1. The GTP agent 210 uses the intelligent selection engine 220 to process the create session request. The intelligent selection engine 220 extracts the APN name from the create session request and determines whether the APN name matches the key 1614 associated with the workflow analyzer 1604. As described above, the workflow analyzers 1604 are associated with the keys 1614, respectively. In this example, it is assumed that the workflow analyzer 1604 has a workflow analyzer associated with the keys 1614 of "Internet 1", "Internet 2 (internet2)" and "Internet 3 (internet3)". After extracting the APN name ("Internet 1") from the create session request 1712, the intelligent selection engine 220 selects the workflow analyzer with the matching key 1614 ("Internet 1"). If there is more than one workflow analyzer 1604 with a matching key 1614, the intelligent selection engine 220 assigns the workflow analyzer with the highest priority 1612. Examples of matching workflow analyzers are as follows:
[0218] Analyzer-1
[0219] Type = Workflow Profile
[0220] key apn name internet 1
[0221] Motion Control Profile WCP1
[0222] Motion Data Profile WDP1
[0223] As shown above, Analyzer 1, which is a workflow analyzer (i.e., "Type = Workflow Profile"), has a key apn name internet1. Analyzer 1 is also associated with an action control profile and an action data profile. The action and control profile information are assigned to the subscriber's session to create a modified create session request 1713. The control profile contains control information associated with the create session request (e.g., which servers to contact, changes in APN names). The data profile is data associated with the control information. In this example, the control profile WCP1 is associated with "OCS1" and "OFCS1", which identify the servers, and the data profile is associated with "Billing Profile 1", which indicates the type of policy to be applied at the identified server.
[0224] An example of a control profile mapping a create session request 1712 to different APNs is shown below: Control profile WCP1
[0225] Mapped apn name new-apn
[0226] OCS interface OCSCHRG
[0227] OFCS interface OFCSCHRG
[0228] Mapped apn report type
[0229] Allow sub-analyzer based on mapped apn (allow-mapped-apn-based-sub-analyzer) error
[0230] As shown above, the control configuration file WCP1 includes a row indicating a mapped apn name ("new apn"). New apn refers to an apn name to which the received apn name will be mapped.
[0231] The OCS interface OCSHRG refers to an online charging system interface used for online charging, quota management, credit control, etc.
[0232] The OFCS interface OFCSCHRG refers to an offline charging server interface to which offline billing records will be transmitted.
[0233] Mapped APN Report Type is a field that relates to which APN name (original or mapped) will be sent to an external server (such as OFCS).
[0234] Allowing sub-parser errors based on mapped APNs involves intelligently choosing whether the engine runs again based on the new APN name received after mapping the original APN.
[0235] The modified create session request 1713 is then sent to the peer selector 804, where the peer node is selected. In this example, the intelligent selection engine 220 uses the matching service network MCC (e.g., 310) to find the peer selector 804. An exemplary matching peer selector 804 is shown below:
[0236] Analyzer-3
[0237] type=gtp peer selector
[0238] Key Service Network MCC 310
[0239] gtp peer list listl (PGW-1, PGW-2, PGW-3)
[0240] As shown above, analyzer 3 has a serving network mcc key of 310, which matches the serving network MCC of the create session request 1712, 1713. Analyzer 3 is associated with the GTP peer list "List 1", which includes PGW-1, PGW-2 and PGW-3. One of PGW-1, PGW-2 and PGW-3 is selected based on the availability and load of the node or a polling algorithm. In the above Figures 8 to 10 Additional details regarding selection between PGW nodes are shown and described in and accompanying text.
[0241] Referring to step 1714, in some embodiments, the GTP agent uses Fig.16 The TEID switching is performed using the TEID mapping table described in , to replace the TEID in the GTP header with the TEID of the selected PGW. The GTP proxy sends its own TEID in the sender FTEID TEID field.
[0242] Referring to step 1716 , a create session request 1816 is transmitted to the selected PGW (eg, PGW1 ).
[0243] Referring to step 1718, after receiving the create session response from PGW1, GTP proxy 210 sends the message to the TEID switching module to replace the TEID in the GTP header with the TEID of the SGW associated with the create session request.
[0244] Referring to step 1720, the packet then passes through the quota management module 1702, where the GTP agent 210 sends subscriber information to the OCS server 1112 to request the quota for the subscriber 1722. In some embodiments, there may be multiple OCS servers in the network. In the example provided above, the quota management module 1702 exchanges messages with the OCS server 1112 ("OCS1"), which is the server to which the workflow control profile WCP1 is directed.
[0245] Referring to step 1724, after successfully receiving the quota, the packet then passes through the CDR generation module 1704, which generates an offline billing record and sends it to the OFCS server 1114 (1726). In the example provided above, the CDR generation engine 1704 then communicates with the OFCS server OFCS1 (1726) to generate the billing record. OFCS1 is the server to which the workflow control profile WCP1 points.
[0246] Referring to step 1728, the Create Session Response is then forwarded to the SGW or MME.
[0247] The GTP agent 210 with the intelligent selection engine 220 can also be used for path management. The GTP agent 210 can initiate and support the GTP-C path management process and the GTP-U path management process on the access interface and the core interface for the peer nodes that are locally configured and dynamically learned. The dynamic peer node can be learned after receiving the create session / create PDP request, and if so configured, the response and path management process can be initiated. The node running the GTP agent service can also maintain its own recovery counter for path management. If the GTP agent 220 detects that the restart counter of the SGW has changed, it can run a cleanup session on the PGW corresponding to the failed SGW. If the GTP agent detects that the restart counter of the PGW has changed, it can run a cleanup session on the SGW / SGSN corresponding to the failed PGW.
[0248] The GTP proxy 210 with the intelligent selection engine 220 can also be used by the operator to perform offline software upgrades of PGW nodes without the need to modify DNS settings on the MME side, because the GTP proxy can select an alternative PGW node.
[0249] Fig.18 is a schematic diagram illustrating path management for echo request / response recovery according to some embodiments of the present disclosure. Fig.18GTP proxy 1910 having an IP address of 10.10.10.10., SGW 1804 having an IP address of 10.11.12.15, and PGW 1808 having an IP address of 10.10.12.13 are shown. In addition to having assigned IP addresses, SGW 1804, GTP proxy 1810, PGW 1808 also have the same functions and structures as SGW 734, GTP proxy 210, and PGW 738, respectively. Fig.18 Also shown are echo request recovery 1822 , echo request recovery 1824 , echo response recovery 1826 , echo response recovery 1828 , and echo request recovery 1832 , echo request recovery 1834 , echo response recovery 1836 , and echo response recovery 1838 .
[0250] SGW 1804 sends an echo request recovery 1822 to the GTP proxy 1810. The echo request is used by a node to ping or recover a connection with a peer node over a GTP interface. The echo request recovery is associated with a value of 10. A GTP entity (e.g., a PGW or SGW or a GTP proxy) may maintain two restart counters. One restart counter is the counter sent by the peer node, and the other restart counter is its own counter. When an entity restarts, it may immediately increment its restart counter and send a new value in a subsequent echo request or response. This value refers to the restart counter (or recovery value) associated with the sending node.
[0251] GTP Proxy 1810 sends an Echo Request Reply 1824 to PGW 1808. The Echo Request Reply is associated with a value of 20.
[0252] PGW 1808 sends an echo response reply 1826 to GTP proxy 1810. The echo response reply is associated with a value of 30.
[0253] GTP proxy 1810 sends an echo response reply 1828 to SGW 1804. The echo response reply is associated with a value of 20.
[0254] PGW 1808 sends an Echo Request Reply 1832 to GTP Proxy 1910. The Echo Request Reply is associated with a value of 30.
[0255] GTP proxy 1810 sends an Echo Request Reply 1834 to SGW 1804. The Echo Request Reply is associated with a value of 20.
[0256] SGW 1804 sends an echo response reply 1836 to GTP proxy 1810. The echo response reply is associated with a value of 10.
[0257] GTP proxy 1810 sends an echo response reply 1838 to PGW 1808. The echo response reply is associated with a value of 20.
[0258] Having the GTP proxy 1810 inserted between the SGW 1804 and the PGW 1808 allows for improved echo messaging. When the GTP proxy is inserted between the SGW and the PGW, it utilizes the SGW and the PGW to handle the path management process. The SGW only sends echo requests and responses to the GTP proxy and is unaware of the presence of the PGW. As a result, the SGW does not have to utilize multiple PGWs in the network to handle the path management process. The PGW also does not have to handle echo requests and responses to / from multiple SGWs, and it is only connected to the GTP proxy interface. This greatly reduces the number of messages exchanged. When the GTP proxy detects that the PGW is down, it can select a different available PGW during session establishment. This is in contrast to the prior art, in which call establishment fails when the SGW cannot reach the PGW.
[0259] Fig.19 is a schematic diagram illustrating path management for PGW failure according to some embodiments of the present disclosure. Fig.19 Shown are GTP-proxy 1910 having an IP address of 10.10.10.10., SGW 1902 having an IP address of 10.11.12.15, SGSN 1904 having an IP address of 10.11.12.16, and PGW 1908 having an IP address of 10.10.12.13. SGW 1902, SGSN 1904, GTP-proxy 1910, PGW 1908 have the same functions and structures as SGW 734, SGSN 732, GTP-proxy 210, and PGW 738, respectively, except for having assigned IP addresses. Fig.19 Also shown are a delete bearer request / response 1922, an echo request recovery 1912, an echo response recovery 1914, and a delete PDP context request / response 1932.
[0260] GTP proxy 1910 transmits echo request resume 1912 to PGW 1908. As described above, echo request is used by a node to ping or resume a connection with a peer node over a GTP interface. Echo request resume is associated with a value of 20.
[0261] PGW 1908 transmits echo response recovery 1914 to GTP agent 1910. Echo response recovery is associated with a value of 31. This value indicates that PGW 1908 has failed. As described above, a GTP entity (e.g., PGW or SGW or GTP agent) can maintain two restart counters. One restart counter is a counter sent by a peer node, and the other restart counter is its own counter. When an entity restarts, it can immediately increment its restart counter and send a new value in a subsequent echo request or response. If the peer node restarts, it sends a new value to the entity. When the entity detects that the value is different compared to what was sent before, it can detect that the peer node has restarted. In this example, if the peer node has previously sent 30, receiving 31 indicates a failure.
[0262] GTP proxy 1910 communicates the failure to SGW 1902 and SGSN 1904 by transmitting a delete bearer request 1912 and receiving a delete bearer response from SGW 1902 and SGSN 1904. GTP proxy 1910 also communicates the failure to SGW 1902 and SGSN 1904 by transmitting a pdp context request to the SGSN and receiving a delete pdpd context response from the SGSN.
[0263] Having the GTP proxy 1910 inserted between the SGWs 1902, 1904 and the PGW 1908 allows for improved handling of failed PGWs. As previously described, when the GTP proxy is inserted between the SGW and the PGW, it utilizes the SGW and the PGW to handle the path management process. The SGW only sends echo requests and responses to the GTP proxy and is unaware of the presence of the PGW. As a result, the SGW does not have to utilize multiple PGWs in the network to handle the path management process. The PGW also does not have to handle echo requests and responses to / from multiple SGWs, and it is only connected to the GTP proxy interface. This greatly reduces the number of messages exchanged. When the GTP proxy detects that the PGW is down, it can select a different available PGW during session establishment. This is in contrast to the prior art, in which call establishment fails when the SGW cannot reach the PGW.
[0264] Fig. 20 is a schematic diagram illustrating path management for SGW failure according to some embodiments of the present disclosure. Fig. 20A GTP proxy 2010 having an IP address of 10.10.10.10, an SGW1 2002 having an IP address of 10.11.12.15, an SGW2 2004 having an IP address of 10.11.12.16, and a PGW 2008 having an IP address of 10.10.12.13 are shown. SGW1 2002 / SGW2 2004, GTP proxy 2010, and PGW 2008 have the same functions and structures as SGW 734, GTP proxy 2010, and PGW 738, respectively, except for having assigned IP addresses. Fig. 20 Also shown are an echo request recovery 2012, a delete session request / response for a session belonging to SGW12014, an echo request recovery 2022, and an echo response recovery 2024.
[0265] The GTP proxy 2010 sends an Echo Request Resume 2012 to SGW1 2012. The Echo Request Resume is associated with a value of 20. Since SGW1 2012 has failed, the GTP proxy 2010 does not receive an Echo Request Response from SGW1 2012.
[0266] The GTP proxy 2010 then sends a Delete Session Request 2014 to the PGW 2008 to delete the session belonging to SGW1 2002. The PGW 2008 confirms in a Delete Session Response 2014 that it has deleted the session belonging to SGW1.
[0267] The GTP proxy 2010 then transmits an Echo Request Reply 2022 to SGW2 2014. The Echo Request Reply 2022 is associated with a value of 20.
[0268] SGW 2004 transmits Echo Response Reply 2024 to GTP Proxy 2010. Echo Response Reply 2024 is associated with a value of 10.
[0269] Having the GTP proxy 2010 inserted between the SGWs 2002, 2004 and the PGW 2008 allows for improved handling of failed SGWs. As previously described, when the GTP proxy is inserted between the SGW and the PGW, it utilizes the SGW and the PGW to handle the path management process. The SGW only sends echo requests and responses to the GTP proxy and is unaware of the presence of the PGW. As a result, the SGW does not have to utilize multiple PGWs in the network to handle the path management process. The PGW also does not have to handle echo requests and responses to / from multiple SGWs, and it is only connected to the GTP proxy interface. This greatly reduces the number of messages exchanged. When the GTP proxy detects that the PGW is down, it can select a different available PGW during session establishment. This is in contrast to the prior art, in which call establishment fails when the SGW cannot reach the PGW.
[0270] Fig.21 21 is a schematic diagram showing an exemplary 5G architecture 2100 according to some embodiments. The 5G architecture 2100 includes a UE 2102, a next generation NodeB ("gNodeB") 2104, an access and mobility management function ("AMF") 2121, a network function ("NF") repository function ("NRF") 2123, a session management function ("SMF") 2125, a user plane function ("UPF") 2127, a unified data management ("UDM") 2131, a PCF 2129, a Gi network 2130, and one or more intelligent selection engines in the intelligent selection engines 2111, 2113, and / or 2115. The gNodeB 2104 can be, for example, a 5G new radio ("NR") base station. The intelligent selection engine can be inserted into various network elements in the 5G core network. For example, in some embodiments, one or more of the smart selection engines 2111, 2113, and / or 2115 may be implemented in AMF 2121, NRF 2123, and 2125, respectively. In an embodiment, the smart selection engine 2113 is inserted into the NRF 2123 and may be used to select various network functions such as AMF 2121, SMF 2125, UPF 2127, UDM 2131, and / or PCF 2129 based on parameters such as, but not limited to, plmnid, data network name ("dnn"), single network slice selection assistance information ("s-nssai"), and tracking areas. The smart selection engine 2113 may run as a microservice or library in the NRF 2123. In some embodiments, the smart selection engine 2113 may run as a container or virtual machine within the NRF 2123 or on a blade. The smart selection engine 2113 may operate similarly to the other smart selection engines described above.
[0271] In some embodiments, in the absence of NRF 2123, the smart selection engine 2115 can be used by SMF 2125 to select UPF 2127, PCF 2129 and / or UDM 2131 for a session using session parameters (such as but not limited to: dnn, s-nssai and tracking areas). The smart selection engine 2115 can run as a microservice or library in SMF 2125. In some embodiments, the smart selection engine 2115 can run as a container or virtual machine within SMF 2125 or on a blade. The smart selection engine 2115 can operate similarly to the other smart selection engines described above.
[0272] In some embodiments, in the absence of NRF 2123, the intelligent selection engine 2111 can be inserted into AMF 2121 and can be used by AMF 2121 to select SMF 2125 for the session using session parameters (dnn, s-nssai and subscriber tracking area). The intelligent selection engine 2111 can run as a microservice or library in AMF 2121. In some embodiments, the intelligent selection engine 2111 can run as a container or virtual machine within AMF 2121 or on a blade. The intelligent selection engine 2111 can operate similarly to the other intelligent selection engines described above.
[0273] Each of the smart selection engines 2111, 2113, 2115 can be deployed independently or in conjunction with another smart selection engine. It should be appreciated that further implementations of the smart selection engine are contemplated within the 5G network.
[0274] The subject matter described herein can be implemented in a digital electronic circuit system, or in computer software, firmware or hardware (including the structural components disclosed in this specification and their structural equivalents), or in a combination thereof. The subject matter described herein can be implemented as one or more computer program products, such as being tangibly implemented in an information carrier (e.g., in a machine-readable storage device) or being implemented in a propagated signal so as to be executed by a data processing device (e.g., a programmable processor, a computer or a plurality of computers) or to control one or more computer programs for the operation of these data processing devices. Computer programs (also referred to as programs, software, software applications or codes) can be written in any form of programming language (including compiled languages or interpreted languages), and computer programs can be deployed in any form, including independent programs or modules, components, subroutines or other units suitable for use in a computing environment. Computer programs do not necessarily correspond to files. Programs can be stored in a portion of a file that stores other programs or data, or in a single file dedicated to the program under discussion, or in multiple collaborative files (e.g., files storing one or more modules, subroutines or partial codes). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.
[0275] The processes and logic flows described in this specification (including the method steps of the subject matter described herein) can be performed by one or more programmable processors that execute one or more computer programs to perform the functions of the subject matter described herein by operating on input data and generating output. The processes and logic flows can also be performed by the devices of the subject matter described herein, and these devices can be implemented as special-purpose logic circuit systems, such as FPGAs (field programmable gate arrays) or ASICs (application-specific integrated circuits).
[0276] Processors suitable for the execution of computer programs include, by way of example, general-purpose microprocessors and special-purpose microprocessors and any one or more processors of any kind of digital computer. In general, the processor will receive instructions and data from a read-only memory or a random access memory or both. The essential elements of a computer are: a processor for executing instructions and one or more memory devices for storing instructions and data. In general, the computer will also include one or more mass storage devices (such as magnetic disks, magneto-optical disks or optical disks) for storing data, or the computer will also be operatively coupled to receive data from the one or more mass storage devices or transfer data to the one or more mass storage devices or both. Information carriers suitable for implementing computer program instructions and data include all forms of non-volatile memory, including, by way of example, semiconductor memory devices (such as EPROM, EEPROM and flash memory devices); magnetic disks (such as internal hard disks or removable disks); magneto-optical disks; and optical disks (such as CD disks and DVD disks). The processor and memory can be supplemented by or incorporated into a dedicated logic circuit system.
[0277] In order to provide interaction with a user, the subject matter described herein can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user. For example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback), and input from the user can be received in any form, including voice input, speech input, or tactile input.
[0278] The subject matter described herein can be implemented in a computing system that includes a back-end component (e.g., a data server), a middleware component (e.g., an application server), or a front-end component (e.g., a client computer with a graphical user interface or a web browser through which a user can interact with an implementation of the subject matter described herein), or any combination of such back-end components, middleware components, and front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include local area networks ("LANs") and wide area networks ("WANs"), such as the Internet.
[0279] It should be understood that the disclosed subject matter is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the accompanying drawings. The disclosed subject matter is capable of other embodiments and can be practiced and implemented in various ways. Moreover, it should be understood that the words and terminology employed herein are for descriptive purposes and should not be considered limiting.
[0280] Thus, those skilled in the art will appreciate that the concepts upon which the present disclosure is based can be readily utilized as a basis for designing other structures, methods, and systems for implementing the multiple purposes of the disclosed subject matter. It is therefore important that the claims be deemed to include such equivalent constructions to the extent that such equivalent constructions do not depart from the spirit and scope of the disclosed subject matter.
[0281] While the disclosed subject matter has been described and illustrated in the foregoing illustrative embodiments, it should be understood that the present disclosure is presented by way of example only and that numerous changes may be made in the details of the implementation of the disclosed subject matter without departing from the spirit and scope of the disclosed subject matter, which is limited only by the claims that follow.
Claims
1. A method for managing a session, comprising: receiving, by a selection engine, a trigger to select a peer node for a subscriber session; selecting, by the selection engine in response to the trigger, a peer selector from among one or more available peer selectors, wherein selecting the peer selector comprises: analyzing subscriber session data to determine one or more parameters of the subscriber session; identifying one or more matching peer selectors among the one or more available peer selectors having a key type that matches the one or more parameters; and selecting the peer selector from the one or more matching peer selectors based on: the selected peer selector having a highest priority among the one or more matching peer selectors; determining, by the selection engine, whether the selected peer selector is associated with a fully qualified domain name (FQDN); In response to determining that the selected peer selector is associated with the FQDN, determining, by the selection engine, whether a Domain Name System (DNS) server is configured, wherein the DNS server is configured when at least one of a primary DNS server or a secondary DNS server is provisioned and reachable; In response to determining that the DNS server is configured, requesting, by the selection engine, from the DNS server a peer list associated with the FQDN; and The peer node is selected from the peer list by the selection engine.
2. The method of claim 1, wherein the trigger comprises at least one of: an indication of session-related control traffic traversing nodes associated with the selection engine, or a request to select the peer node.
3. The method according to claim 1, further comprising: The peer list is obtained, by the selection engine, from the selected peer selector in response to determining that (a) the selected peer selector is not associated with the FQDN or (b) the DNS server is not configured.
4. The method of claim 1 , wherein selecting the peer node from the peer list further comprises: receiving, by the selection engine, the peer list; excluding, by the selection engine, one or more unreachable peer nodes from the peer list; excluding, by the selection engine, one or more overloaded peer nodes from the peer list; grouping, by the selection engine, the one or more remaining peer nodes in the peer list based on a load level associated with each of the one or more remaining peer nodes to generate a plurality of peer groups each including one or more peer nodes; as well as The peer node is selected by the selection engine from the least loaded peer group among the plurality of peer groups.
5. The method of claim 4, wherein each of the one or more unreachable peer nodes is at least one of: out of order, under maintenance, or disconnected from the network.
6. The method of claim 4, wherein each of the one or more overloaded peer nodes is at least one of: at maximum capacity, within a threshold of reaching maximum capacity, or experiencing issues associated with software corruption.
7. The method according to claim 4, further comprising: A status message is received by the selection engine, the status message identifying at least one of: the one or more unreachable peer nodes or the one or more overloaded peer nodes.
8. The method of claim 4, wherein the peer node selects from the least loaded peer group using at least one of round robin, weighted round robin, or hashing.
9. The method of claim 1, wherein the selection engine is implemented in one of: a mobility management engine for selecting a serving gateway peer node; An evolved packet data gateway for selecting a packet data network gateway peer node; A control plane node for selecting a user plane peer node; or GPRS Tunneling Protocol Proxy GTP Proxy used to select a gateway peer node.
10. A system for managing a session, comprising: Memory; as well as One or more processors are coupled to the memory, the one or more processors being configured to read instructions from the memory, the instructions causing the one or more processors to perform operations during execution, the operations comprising: receiving a trigger to select a peer node for a subscriber session; In response to the trigger, selecting a peer selector from one or more available peer selectors, wherein selecting the peer selector comprises: Analyze subscriber session data to determine one or more parameters of the subscriber session number; identifying one or more matching peer selectors among the one or more available peer selectors having a key type that matches the one or more parameters; and selecting the peer selector from the one or more matching peer selectors based on: the selected peer selector having a highest priority among the one or more matching peer selectors; determining whether the selected peer selector is associated with a fully qualified domain name (FQDN); In response to determining that the selected peer selector is associated with the FQDN, determining whether a Domain Name System (DNS) server is configured, wherein the DNS server is configured when at least one of a primary DNS server or a secondary DNS server is provisioned and reachable; In response to determining that the DNS server is configured, requesting a peer list associated with the FQDN from the DNS server; and The peer node is selected from the peer list by a selection engine.
11. The system according to claim 10, further comprising: The peer list is obtained, by the selection engine, from the selected peer selector in response to determining that (a) the selected peer selector is not associated with the FQDN or (b) the DNS server is not configured.
12. The system of claim 10, wherein selecting the peer node from the peer list further comprises: receiving, by the selection engine, the peer list; excluding one or more unreachable peer nodes from the peer list; excluding one or more overloaded peer nodes from the peer list; grouping the one or more remaining peer nodes in the peer list based on a load level associated with each of the one or more remaining peer nodes to generate a plurality of peer groups each including the one or more peer nodes; as well as The peer node is selected from a peer group having the smallest load among the multiple peer groups.
13. A non-transitory computer-readable medium storing instructions that, when executed by a selection engine comprising one or more hardware processors, cause the selection engine to perform operations comprising: receiving a trigger to select a peer node for a subscriber session; In response to the trigger, selecting a peer selector from one or more available peer selectors, wherein selecting the peer selector comprises: analyzing subscriber session data to determine one or more parameters of the subscriber session; identifying one or more matching peer selectors among the one or more available peer selectors having a key type that matches the one or more parameters; and selecting the peer selector from the one or more matching peer selectors based on: the selected peer selector having a highest priority among the one or more matching peer selectors; determining whether the selected peer selector is associated with a fully qualified domain name (FQDN); In response to determining that the selected peer selector is associated with the FQDN, determining whether a Domain Name System (DNS) server is configured, wherein the DNS server is configured when at least one of a primary DNS server or a secondary DNS server is provisioned and reachable; In response to determining that the DNS server is configured, requesting a peer list associated with the FQDN from the DNS server; and The peer node is selected from the peer list by the selection engine.
14. The non-transitory computer-readable medium of claim 13, wherein the trigger comprises at least one of: an indication of session traffic traversing nodes associated with the selection engine, or a request to select the peer node.
15. The non-transitory computer readable medium of claim 13, further comprising: The peer list is obtained, by the selection engine, from the selected peer selector in response to determining that (a) the selected peer selector is not associated with the FQDN or (b) the DNS server is not configured.
16. The non-transitory computer-readable medium of claim 13, wherein selecting the peer node from the peer list further comprises: receiving the peer list; excluding one or more unreachable peer nodes from the peer list; excluding one or more overloaded peer nodes from the peer list; grouping the one or more remaining peer nodes in the peer list based on a load level associated with each of the one or more remaining peer nodes to generate a plurality of peer groups each including the one or more peer nodes; as well as A peer node is selected from a peer group having the smallest load among the plurality of peer groups.
Citation Information
Patent Citations
Adapted television antenna and associated amplifier for vehicles
EP0000204A1
Methods and apparatus for relocating and restoring connections through a failed serving gateway and traffic offloading
CN103202064A
Node selection in a packet core network
CN103262503A
Method and system for redirecting data requests in peer-to-peer data networks
US20030050966A1
Method and Apparatus for Pooling Network Resources
US20100291943A1
