Hierarchical caching in telecommunication networks

• US 7,860,948 B2
• Filed: 09/03/2002
• Issued: 12/28/2010
• Est. Priority Date: 09/26/2001
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Hameleers, Heino, Hundscheidt, Frank
• Primary Examiner(s): Nguyen; Dustin
• Application Number:
  US10/490,841
• Publication Number:
  US 20040255016A1
• Time in Patent Office: 3,038 Days
• Field of Search: 709/203, 709/216, 709/217, 455/456.5, 707/999.102
• US Class Current: 709/219
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 67/2885 Hierarchically arranged intermediate devices, e.g. for hierarchical caching
• CPC Class Codes: H04L 67/568 Storing data temporarily at an intermediate stage, e.g. caching
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

The invention addresses the field of mobile communications. It relates to a method, network node and computer program for management of cached information within a hierarchical structure. The basic approach is to provide an intelligent mechanism for more dynamic management of the cached contents. In particular it provides a solution for a more efficient utilisation of the storage resources in a caching architecture by caching a single cached copy of the same content on each path from a client to the original source in the hierarchical architecture. This basically means that the cached information is placed as high up the hierarchy as necessary, but that the mechanism is enhanced with the capability to also move the cached information down the hierarchy in case the content is mainly used on a regular base by hosts connected to a part of the hierarchical architecture. The dynamics of the proposed solution foresees also to replicate a cached content to more then one part of the hierarchical architecture.

Claims

1. A method for dynamically managing cached information in a communication network with network nodes related in a hierarchical architecture with at least one server on the top, at least one host at the bottom of the hierarchical architecture and with at least one cache engine for caching the information between the at least one server and the at least one host, comprising:

establishing access statistics for the cached information at a location within the hierarchical architecture;

responsive to receiving a request for the cached information from the at least one host, checking the access statistics of at least one other location; and

evaluating the access statistics of both locations to determine the location of the cached information relative to the at least one host and nearby hosts;

optimizing access for the at least one host and nearby hosts by moving the cached information up or down within the hierarchical architecture;

accessing the cached information by

the at least one host sending a request message for information towards at least one server;

the next passed cache engine checking if it has the requested information;

the requested information, being available, is returned and the access statistics of the information by other hosts is checked, whereupon a decision is taken and the next passed cache engine having a pointer to the location of the requested information, forwarding the request message to the location of the requested information, else the requested message being forwarded to a next cache engine and

retrieving the information from the location closest to the at least one host.

2. The method according to claim 1 wherein the access statistics are evaluated either in the cache engine or in the at least one server or in a router, which is responsible for routing information within the hierarchical architecture.

3. The method according to claim 1, wherein the cached information is replicated within the hierarchical architecture.

4. The method according to claim 1, in case the determination of the location results in changing the location, the cache engine stores a pointer to the location of the information.

5. The method according to claim 1, wherein the access statistics consider the number of accesses per time unit and the number of hosts for a content.

6. The method according to claim 1, wherein the access statistics consider distance and priority of the locations.

7. The method according to claim 1, wherein a cache engine closest to the host adds to a request message for information sent from the host towards the at least one server an indication indicating the wish for caching the information.

8. The method according to claim 7, wherein a next passed cache engine, which is passed by the request message on the way towards the at least one server, overrules the indication.

9. The method according to claim 1, wherein the pointer to the location of the information is accompanied with any additional information from a group comprising caching time, cache refresh rate and a distance indication to the cache engine being located lower in the hierarchical architecture.

10. The method according to claim 1, wherein in case a cache engine has the pointer to the location of the information, the distance to the cache engines having the cached information is compared with the distance to the at least one server and in case the distance to the at least one server is shorter then the information is fetched from the at least one server.

11. The method according to claim 1, wherein the cached information includes an information indicating the likelihood that the cached content is changed within a certain time period.

12. The method according to claim 1, wherein in order to update the cached information a refresh mechanism is carried out.

13. The method according to claim 12, wherein the refresh mechanism is a push-based mechanism initiated by the at least one server.

14. The method according to claim 13, wherein the at least one server sends an updated version of the content along the hierarchical architecture by taking into consideration the pointers of the location of the information.

15. The method according to claim 12, wherein the refresh mechanism is a pull-based mechanism initiated by at least one cache engine.

16. The method according to claim 1, wherein an information indicating the likelihood that the cached content is changed within a certain time period is considered.

17. The method according to claim 16, wherein sending an additional message to the cache engine or automatically when it is outdated does a removal of the cached information.

18. The method according to claim 17, wherein the establishment of the hierarchical architecture is done with a multicast and a corresponding multicast delivery tree topology.

19. The method according to claim 18, wherein the establishment of the hierarchical architecture is done with a routing protocol in order to get access to information about the route between the at least one server and the clients.

20. The method according to claim 19, wherein the routing protocol is a distance vector routing protocol containing a table with the shortest paths to all nodes within a network.

21. The method according to claim 1 the communication between two network nodes is based on a point to point protocol.

22. A network node, comprising digital processing unit and memory, adapted to dynamically manage cached information in a communication network with said and further network nodes related in a hierarchical architecture with at least one server on the top, with at least one host at the bottom of the hierarchical architecture and with at least one cache engine for caching the information between the at least one server and the at least one host, the network node comprising:

a collecting means for collecting data about the usage of the cached information within the hierarchical architecture,

a statistics means for determining of access statistics for the cached information from at least one other location;

an analysis means for evaluating the access statistics of both locations to determine to determine the location of the cached information relative to the at least one host and nearby hosts, wherein optimizing the location of the cached information moves the cached information up or down within the hierarchical architecture, and the cached information is accessed by;

the at least one host sending a request message for information towards at least one server;

the next passed cache engine checking if it has the requested information;

the requested information, being available, is returned and the access statistics of the information by other hosts connected to the next passed engine is checked whereupon a decision is taken and the next passed cache engine having a pointer to the location of the requested information, forwarding the request message to the location of the requested information, else the request message being forwarded to a next cache engine; and

a retrieval means for retrieving the information from the location closest to the at least one host.

23. The network node according to claim 22, wherein the network node is one of the server, the cache engine and a router for routing purposes within the hierarchical architecture.

24. A system comprising a digital processing unit and memory and adapted to dynamically manage cached information in a communication network with network nodes related in a hierarchical architecture with at least one server on the top, with at least one host at the bottom of the hierarchical architecture and with at least one cache engine for caching the information between the at least one server and the at least one host, and the system includes network nodes, wherein each network node comprises:

a collecting means for collecting data about the usage of the cached information within the hierarchical architecture;

a statistics means for determining access statistics for the cached information from the collected data and

an analysis means for evaluating the access statistics of both locations to determine the location of the cached information relative to the at least one host and nearby hosts, wherein the optimization of the location of the cached information results in moving the cached information up or down within the hierarchical architecture, and wherein in order to access the cached information the following steps are to be performed;

the at least one host sends the'"'"' request message for information towards the at least one server,

the next passed cache engine checks if it has the requested information,

in case of having the information available, this information is returned and the access statistics of the information by other hosts connected to this cache engine is checked, and

considering the result a decision about the location of the information is taken, and

in case the next passed cache engine has the pointer to the location of the information, the request message is forwarded to the location of the information, else the request message is forwarded to the next cache engine; and

retrieval means for retrieving the information from the location closest to the at least one host.

25. The system according to claim 24 wherein the hierarchical architecture has a form of a tree with at least one server on the top of the tree, with at least one host as a leaf of the tree, and in-between with at least one router and with at least one cache engine, which stores cached information and is connected to a corresponding router.