Arrangement for load sharing in computer networks

• US 6,687,731 B1
• Filed: 02/07/2000
• Issued: 02/03/2004
• Est. Priority Date: 06/12/1997
• Status: Expired

• Current Assignee: Hanger Solutions, LLC (IP Investments Group LLC)
• Original Assignee: Telia AB (Government of Norway)
• Inventors: Kavak, Nail
• Primary Examiner(s): Jean, Frantz B
• Application Number:
  US09/424,872
• Time in Patent Office: 1,457 Days
• Field of Search: 709/105, 709/203, 709/219, 709/242, 709/238, 709/226, 370/389
• US Class Current: 718/105
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/101 based on network conditions
• CPC Class Codes: H04L 67/1021 based on client or server locations
• CPC Class Codes: H04L 67/1036 Load balancing of requests to servers for services different from user content provisioning, e.g. load balancing across domain name servers
• CPC Class Codes: H04L 67/1038 Load balancing arrangements to avoid a single path through a load balancer
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

The invention relates to an arrangement for load sharing in computer networks and, more exactly, an arrangement for distribution of traffic, for instance via Internet, from clients (1) to service suppliers who provide services from many servers. The invention makes possible distribution to one of a number of replicated servers. Suitable server is selected for instance on basis of available resources at the interface of the server, or less delay in the transmission. The invention results in better performance and reduced traffic by distribution of the traffic geographically and from a resource point of view. According to the invention, a number of replicated servers (5A-5E) belong to an anycast-group and each anycast-group is connected to a domain name server (2) which has the ability to select one of the replicated servers, so that a router (4) can establish a connection between the selected server and the service-requesting client computer. Each replicated server (5A-5E) can transmit a resource advertisement which contains information about available resources at the server in question, and about the link parameters of the server.

Claims

1. A computer implemented system for designating a resource location among a plurality of resource locations, comprising:

at least one client computer;

at least one service supplier responsive to requests of the at least one client computer, the at least one service supplier comprising,

i. a domain server configured to provide access to the plurality of resource locations in response to a request from the at least one client computer, the plurality of resource locations being replicated servers operably linked to the domain server, a portion of the replicated servers being configured to be responsive to an anycast address protocol,

ii. at least one router responsive to the portion of replicated servers, the router having a routing table configured to receive resource advertisements from the portion of replicated servers, the resource advertisements including at least capacity data of each replicated server, the router being operably linked to the domain server for resolving the designation of the resource location among the plurality of resource locations,

wherein the domain server utilizes the anycast address protocol for accessing the plurality of resource locations, the router table being responsive to the anycast requests of the domain server to designate the resource location for access by the at least one client computer in accordance with the resource advertisements of the table, and

wherein a resource advertisement is periodically transmitted from a corresponding replicated server upon a change of a predetermined amount in the capacity data of the corresponding replicated server.

2. The computer implemented system of claim 1, wherein the at least one router designates the resource location as the replicated server having the most available capacity in accordance with the resource advertisements of the table.

3. The computer implemented system of claim 1, wherein the resource advertisement includes link data.

4. The computer implemented system of claim 3, wherein the link data includes parameters for indicating a link latency value.

5. The computer implemented system of claim 1, wherein the routing table designates the resource location based only on a location of each replicated server.

6. The computer implemented system of claim 1, wherein the resource advertisement includes at least one of token bucket depth and token bucket rate.

7. The system of claim 1, wherein the replicated servers are arranged in a service hierarchy and respective Private Network Node Interface (PNNI) routing levels are used to identify respective hierarchical levels of the replicated servers.

8. The system of claim 7, wherein the at least one router designates the resource location as a replicated server on a lowest possible level in the service hierarchy, a difference in the respective PNNI routing levels of any two replicated servers in the service hierarchy being used as a measure of proximity between the two replicated servers.

9. A method of designating a replicated server among a plurality of replicated servers, comprising:

receiving, by a domain server, a service request from at least one client computer, the domain server operably linked to the plurality of replicated servers, wherein the replicated servers are accessible using an anycast address protocol;

receiving, by at least one router, an anycast service request from the domain server, each router operably linked to the domain server for resolving the designation of the replicated server among the plurality of replicated servers;

receiving, by a respective routing table of the at least one router, resource advertisements from the plurality of replicated servers, the resource advertisements including respective capacity data of the plurality of replicated servers; and

designating the replicated server among the plurality of replicated servers based on information stored in the respective routing table,

wherein a resource advertisement is periodically transmitted from each replicated server upon a change of a predetermined amount in the capacity data of the corresponding replicated server.

10. The method of claim 9, wherein the at least one router designates the replicated server having the most available capacity in accordance with the information in the respective routing table.

11. The method of claim 9, wherein the resource advertisements include link data.

12. The method of claim 9, wherein the link data includes parameters indicating a link latency value.

13. The method of claim 9, wherein the routing table designates the resource location based only on a location of each replicated server.

14. The method of claim 9, wherein the resource advertisement includes at least one of token bucket depth and token bucket rate.

15. The method of claim 9, wherein the replicated servers are arranged in a service hierarchy and respective Private Network Node Interface (PNNI) routing levels are used to identify respective hierarchical levels of the replicated servers.

16. The method of claim 15, wherein the at least one router designates the resource location as a replicated server on a lowest possible level in the service hierarchy, a difference in the respective PNNI routing levels of any two replicated servers in the service hierarchy being used as a measure of proximity between the two replicated servers.