Identity-based casting of network addresses

• US 8,996,607 B1
• Filed: 06/04/2010
• Issued: 03/31/2015
• Est. Priority Date: 06/04/2010
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Bajaj, Girish Bansilal, An, Zhi
• Primary Examiner(s): CHOUDHURY, AZIZUL Q
• Application Number:
  US12/801,344
• Time in Patent Office: 1,761 Days
• US Class Current: 709/203
• CPC Class Codes: H04L 67/1006 with static server selection, e.g. the same server being selected for a specific client
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30

Abstract

Systems, methods, and computer-readable media identify a server to a client device. In one implementation, server identification data is stored for a plurality of servers associated with a service. An identifier of a client device is received. One of the plurality of servers is selected to provide the service to the client device, and the server is selected based on the identifier of the client device. The server identification data for the selected server is provided to the client device. The selected server may receive a request to access the service from the client device, and may process the request consistent with state data stored at the selected server.

Claims

1. A computer-implemented method for identifying a server to a client, comprising:

storing, at an allocation server, a first set of server identification data for a plurality of content servers associated with a domain name, the plurality of content servers including at least a first content server and a second content server, wherein the allocation server is configured to store server identification data for multiple pluralities of content servers, each plurality of content servers associated with a different domain name;

receiving, at the allocation server, a first request to access one of the plurality of content servers associated with the domain name from a first client device, the first request including an identifier of the first client device;

receiving, at the allocation server, a second request to access one of the plurality of content servers associated with the domain name from a second client device, the second request including an identifier of the second client device;

selecting the first content server to process the first request, wherein the first content server is selected based on at least the identifier of the first client device and the domain name, and the proximity of a location of a hash value for the first client device on a circle to a location of a hash value for the first content server on the circle, and wherein the location of the hash value for the first client device and the location of the hash value for the first content server on the circle are determined by normalizing the hash value of the first client device and the hash value of the first content server to between 0 and 1;

selecting the second content server to process the second request, wherein the second content server is selected based on at least the identifier of the second client device and the domain name, and the proximity of a location of a hash value for the second client device on the circle to a location of a hash value for the second content server on the circle, and wherein the location of the hash value for the second client device and the location of the hash value for the second content server on the circle are determined by normalizing the hash value of the second client device and the hash value of the second content server to between 0 and 1;

providing the server identification data for the first content server to the first client device, wherein the first client device stores the server identification data for the first content server enabling the first client device to communicate with the first content server on subsequent requests to the plurality of content servers associated with the domain name while bypassing the allocation server; and

providing the server identification data for the second content server to the second client device, wherein the second client device stores the server identification data for the second content server enabling the second client device to communicate with the second content server on subsequent requests to the plurality of content servers associated with the domain name while bypassing the allocation server,

wherein the first content server processes the subsequent requests of the first client device consistent with first state data stored at the first content server, the first state data corresponding to the first client device,

wherein the second content server processes the subsequent requests of the second client device consistent with second state data stored at the second content server, the second state data corresponding to the second client device, and

wherein the first state data includes information related to an interaction between the first client device and the first content server that occurred while processing the first request.

2. The computer-implemented method according to claim 1, further comprising:

receiving an indication that the first content server is unavailable to process subsequent requests from the first client device;

selecting a third content server from the plurality of content servers to handle subsequent requests from the first client device; and

providing the server identification data for the third content server to the first client device.

3. The computer-implemented method according to claim 1, implemented by a domain name system (“

DNS”

) server, wherein the first request and the second request include the domain name.

4. The computer-implemented method according to claim 1, wherein the server identification data for the first content server comprises a first internet protocol (“

IP”

) address and the server identification data for the second content server comprises a second IP address.

5. The computer-implemented method according to claim 4, wherein:

a domain name system (“

DNS”

) server provides the first client device and the second client device with a common internet protocol (“

IP”

) address after receiving corresponding initial requests to access the domain name,

the first client device sends subsequent requests to the first IP address; and

the second client device sends subsequent requests to access the domain name to the second IP address.

6. The computer-implemented method according to claim 4, wherein:

a domain name system (“

DNS”

) server provides the first client device and the second client device with an internet protocol (“

IP”

) address randomly selected from a pool of IP addresses associated with the domain name, the randomly selected address being provided by the DNS server upon receiving corresponding initial requests to access the domain name from the first client device and the second client device,

the first client device sends subsequent requests to access the domain name to the first IP address; and

the second client device sends subsequent requests to access the domain name to the second IP address.

7. A computer-implemented method for identifying a server, comprising:

storing, at an allocation server, a first set of server identification data for a plurality of content servers associated with a service, wherein the allocation server is configured to store server identification data for multiple pluralities of content servers, each plurality of content servers associated with a different service;

receiving an identifier of a client device;

selecting one of the plurality of content servers to provide the service to the client device in response to receiving the identifier, wherein the content server is selected based on at least the identifier of the client device, and the proximity of a location of a hash value for the client device on a circle to a location of a hash value for the selected content server on the circle, and wherein the location of the hash value for the client device and the location of the hash value for the selected content server on the circle are determined by normalizing the hash value of the client device and the hash value of the selected content server to between 0 and 1; and

providing the server identification data for the selected content server to the client device, enabling the client device to store the server identification data for the selected content server and to communicate with the selected content server on subsequent requests to access the service without accessing the allocation server during the subsequent requests,

wherein the selected content server receives a request to access the service from the client device, and stores state data at the selected content server,

wherein the state data includes information related to an interaction between the client device and the selected content server, and

wherein the selected content server processes the subsequent requests consistent with the state data.

8. The computer-implemented method according to claim 7, wherein the state data corresponds to a browsing or connection session between the client device and the selected content server.

9. The computer-implemented method according to claim 8, wherein the state data comprises shopping cart data reflecting items selected for purchase during the browsing or connection session.

10. The computer-implemented method according to claim 7, further comprising:

storing the state data on a backup server selected from the plurality of content servers.

11. The computer-implemented method according to claim 10, further comprising:

receiving an indication that the selected content server is unavailable to process requests from the client device; and

providing the client device with the server identification data for the backup server, such that the subsequent requests from the client device to access the service are received by the backup server.

12. The computer-implemented method according to claim 7, wherein selecting one of the plurality of content servers further comprises:

calculating hash values for the plurality of content servers;

mapping the hash values for the plurality of content servers to locations on the circle;

calculating the hash value for the client device;

mapping the hash value of the client device to the location on the circle.

13. The computer-implemented method according to claim 7, wherein the identifier of the client device comprises a serial number of the client device.

14. The computer-implemented method according to claim 7, wherein the identifier of the client device is received with a Uniform Resource Locator (“

URL”

), and the URL includes a domain name associated with the service.

15. The computer-implemented method according to claim 14, wherein the identifier of the client device is included as a prefix to the domain name associated with the service.

16. A system comprising:

a processor for executing program instructions; and

a non-transitory computer-readable storage medium storing the program instructions, the program instructions, when executed by the processor, performing a process to;

store a first set of server identification data for a first plurality of servers associated with a first service;

store a second set of server identification data for a second plurality of servers associated with a second service;

receive an identifier of a client device and an identifier of a service, wherein the identifier of the service identifies one of the first service and the second service;

determine a requested service based, at least in part, on the received identifier of the service;

select one of a plurality of servers to provide the requested service to the client device, wherein the server is selected based at least on the identifier of the client device and the identifier of the service, and the proximity of a location of a hash value for the client device on a circle to a location of a hash value for the selected server on the circle, and wherein the location of the hash value for the client device and the location of the hash value for the selected server on the circle are determined by normalizing the hash value of the client device and the hash value of the selected server to between 0 and 1; and

provide the server identification data for the selected server to the client device enabling the client device to store the server identification data for the selected server and to communicate with the selected server on subsequent requests to access the service without accessing the system during the subsequent requests,

wherein the selected server receives a request to access the requested service from the client device, and stores state data at the selected server,

wherein the state data includes information related to an interaction between the client device and the selected server that occurred prior to processing the request, and

wherein the selected server processes the subsequent requests consistent with the state data.

17. A non-transitory computer-readable storage medium storing program instructions for performing a method executed by a processor, the method comprising steps performed by the processor of:

storing, at an allocation server, a first set of server identification data for a first plurality of servers associated with a first service;

storing, at the allocation server, a second set of server identification data for a second plurality of servers associated with a second service;

receiving, at the allocation server, a request to access the first service and an identifier of a client device;

selecting, based on the request to access the first service, one of the first plurality of servers to provide the service to the client device, wherein the server is selected based on at least the identifier of the client device, and the proximity of a location of a hash value for the client device on a circle to a location of a hash value for the selected server on the circle, and wherein the location of the hash value for the client device and the location of the hash value for the selected server on the circle are determined by normalizing the hash value of the client device and the hash value of the selected server to between 0 and 1; and

providing the server identification data for the selected server to the client device, enabling the client device to store the server identification data for the selected server and to communicate with the selected server on subsequent requests to access the service without accessing the allocation server during the subsequent requests,

wherein the selected server receives a request to access the service from the client device, and stores state data at the selected server,

wherein the state data includes information related to an interaction between the client device and the selected server, and

wherein the selected server processes the subsequent requests consistent with the state data.

18. A client device comprising:

a processor for executing program instructions; and

a non-transitory computer-readable storage medium storing the program instructions, the program instructions, when executed by the processor, performing a process to request a service comprising;

transmitting a domain name to a first server, the first server comprising a domain name system (“

DNS”

) server;

receiving, from the DNS server, identification data of a second server, the second server comprising an allocation server;

transmitting identification data of the client device to the allocation server, the allocation server configured to store server identification data for multiple pluralities of content servers, each plurality of content servers associated with a different service;

receiving, from the allocation server, identification data of a third server, the third server comprising a content server from the plurality of content servers associated with the service, wherein the third server is determined based at least in part on the proximity of a location of a hash value for the client device on a circle to a location of a hash value for the third server on the circle, and wherein the location of the hash value for the client device and the location of the hash value for the third server on the circle are determined by normalizing the hash value of the client device and the hash value of the third server to between 0 and 1;

requesting the service from the content server, wherein the content server is configured to store state data related to an interaction between the client device and the content server; and

sending a subsequent request to the content server without accessing the allocation server during the subsequent request, wherein the content server is further configured to process the subsequent request consistent with the state data.

19. The client device according to claim 18, wherein the first server and the second server are implemented in the same system.

20. The client device according to claim 18, the process further comprising:

determining whether the DNS server also functions as the allocation server;

when the DNS server also functions as the allocation server, transmitting the domain name and the client identification data together; and

when the DNS server does not function as the allocation server, transmitting the domain name to the DNS server separately from transmitting the client identification data to the allocation server.

21. The client device according to claim 18, wherein:

the identification data of the second server is an IP address of the second server; and

the identification data of the third server is an IP address of the third server.