Integrated adaptive anycast for content distribution

• US 8,560,598 B2
• Filed: 12/22/2009
• Issued: 10/15/2013
• Est. Priority Date: 12/22/2009
• Status: Expired due to Fees

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Santoro, Mario, Brotman, Herani, Glasser, Alan L., Miros, James, Spatscheck, Oliver, Van der Merwe, Jacobus
• Primary Examiner(s): Chan, Wing F
• Assistant Examiner(s): Thieu, Benjamin M
• Application Number:
  US12/645,006
• Publication Number:
  US 20110153719A1
• Time in Patent Office: 1,393 Days
• Field of Search: 709/203, 709/217, 709/219, 709225-227, 709/238, 370/389, 370/392
• US Class Current: 709/203
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 45/126 minimising geographical or physical path length
• CPC Class Codes: H04L 61/4511 using domain name system [DNS]
• CPC Class Codes: H04L 61/5007 Internet protocol [IP] addresses
• 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 67/1021 based on client or server locations

Abstract

A system includes first and second cache servers a domain name server, and a route controller. The cache servers are each configured to respond to an anycast address. Additionally, the first cache server is configured to respond to a first unicast address, and the second cache server is configured to respond to a second unicast address. The router controller configured to determine wither the status of the first cache server is non-overloaded, overloaded, or offline. The route controller is further configured to instruct the domain name server to provide the second unicast address when the status is overloaded or offline, and modify routing of the anycast address to direct a content request sent to the anycast address to the second cache server when the status is offline. The domain name server is configured to receive a request from a requestor for a cache server address. Additionally, the domain name server is configured to provide an anycast address to the requestor when the status of the first cache server is non-overloaded, and provide the second unicast address to the requestor when the status of the first cache server is offline or overloaded.

Claims

1. A system comprising:

first and second cache servers, wherein each of the cache servers responds to an anycast address, wherein the first cache server responds to a first unicast address, and the second cache server responds to a second unicast address;

a domain name server that performs operations comprising;

receiving a request from a requestor for a cache server address;

providing the anycast address to the requestor when a status of the first cache server is non-overloaded; and

providing the second unicast address to the requestor when the status of the first cache server is offline or overloaded; and

a route controller that performs operations comprising;

determining the status of the first cache server, the status being non-overloaded, overloaded, or offline;

instructing the domain name server to provide the second unicast address to the requestor when the status is overloaded or offline; and

modifying routing of the anycast address to direct a content request sent to the anycast address to the second cache server when the status is offline.

2. The system of claim 1 wherein the domain name server further performs operations comprising providing the first anycast address to the requestor when the status of the first cache server is non-overloaded and the requestor is associated with an unstable network.

3. The system of claim 1 wherein the routing of the anycast addresses directs the content request sent to the anycast address to the first cache server when the status is non-overloaded or overloaded.

4. The system of claim 1 wherein the route controller further performs operations comprising receiving a notification from the first cache server.

5. The system of claim 4 wherein the status of the first cache server is offline when a notification from the first cache server is not received.

6. The system of claim 4 wherein the status of the first cache server is overloaded when the notification from the first cache server is received and the load exceeds a threshold.

7. The system of claim 6 wherein the load of the first cache server is based on one of network bandwidth, central processing unit utilization, memory utilization, and number of requests being served for the respective cache server.

8. The system of claim 4 wherein the status of the first cache server is non-overloaded when the notification from the first cache server is received and the load does not exceed a threshold.

9. A domain name server comprising:

a processor that executes instructions to perform operations comprising;

receiving a plurality of address requests from a requestor for a cache address, the requestor being part of a serving region;

identifying a first cache server serving content requests sent to an anycast address from the serving region;

determining a load of the first cache server;

receiving an instruction to provide a unicast address of an alternate cache server to the requestor when the load exceeds an overloaded threshold;

providing, based on the instruction, the unicast address of the alternate cache server to the requestor in response to the address requests when the load exceeds the overloaded threshold;

providing the anycast address in response to the address requests when the load is at or below a pre-overloaded threshold; and

providing the anycast address to the requestor in response to a first portion of the address requests and providing the unicast address to the requestor in response to a second portion of the address requests when the load exceeds the pre-overloaded threshold and is at or below the overloaded threshold.

10. The domain name server of claim 9 wherein the alternate cache servers is selected from a list of cache servers.

11. The domain name server of claim 9 wherein the alternate cache server is based on a load of each of the cache servers.

12. The domain name server of claim 9 wherein the overloaded threshold is at least about 80% of the capacity of the first cache server.

13. The domain name server of claim 9 wherein the pre-overloaded threshold is at least about 60% of the capacity of the first cache server.

14. The domain name server of claim 9 wherein the load of the first cache server is based on one of network bandwidth, central processing unit utilization, memory utilization, and number of requests being served.

15. A computer readable device comprising instructions, which when loaded an executed by a processor, cause the processor to perform operations comprising:

determining the status of a serving region cache server, the status being one of non-overloaded, overloaded, or offline;

instructing a domain name server to provide a unicast address for an alternate cache server when the status is overloaded or offline; and

modifying routing of an anycast address to direct a content request sent to the anycast address to the alternate cache server when the status of the serving region cache server is offline.

16. The computer readable device of claim 15 wherein the serving region includes a plurality of client systems sharing a common ingress point.

17. The computer readable device of claim 15 load information from the serving region cache server.

18. The computer readable device of claim 17 wherein the load information includes one of network bandwidth, central processing unit utilization, memory utilization, and number of requests being served.

19. The computer readable device of claim 15 wherein the operations further comprise identifying the alternate cache server.

20. The computer readable device of claim 19 wherein identifying the alternate cache server comprises determining load information for each of the plurality of cache servers.