Content-aware switching of network packets

• US 6,449,647 B1
• Filed: 09/21/1999
• Issued: 09/10/2002
• Est. Priority Date: 08/01/1997
• Status: Expired due to Term

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Systems, Inc.
• Inventors: Krawczyk, John J., Wasson, Scott, Stevens, Richard C., Royce, Katherine, Colby, Steven, Siegel, Kenneth P., Nair, Rai Krishnan
• Primary Examiner(s): Maung, Zarni
• Application Number:
  US09/400,635
• Time in Patent Office: 1,085 Days
• Field of Search: 709/226, 709/239, 709/240, 709/245, 709/224, 709/250
• US Class Current: 709/226
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/724 at intermediate nodes, e.g. resource reservation protocol [RSVP]
• CPC Class Codes: H04L 47/745 Reaction in network
• CPC Class Codes: H04L 47/801 Real time traffic
• CPC Class Codes: H04L 47/805 QOS or priority aware
• CPC Class Codes: H04L 47/822 Collecting or measuring resource availability data
• CPC Class Codes: H04L 49/355 Application aware switches, e.g. for HTTP
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/10015 Access to distributed or replicated servers, e.g. using brokers
• 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 67/1012 based on compliance of requirements or conditions with available server resources
• CPC Class Codes: H04L 67/1014 based on the content of a request
• CPC Class Codes: H04L 67/1021 based on client or server locations
• CPC Class Codes: H04L 67/1023 based on a hash applied to IP addresses or costs
• CPC Class Codes: H04L 67/1027 Persistence of sessions during load balancing
• CPC Class Codes: H04L 67/1029 using data related to the state of servers by a load balancer
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services
• CPC Class Codes: H04L 67/52 specially adapted for the location of the user terminal
• CPC Class Codes: H04L 69/329 : in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 : Network security protocols

Abstract

A content-aware flow switch intercepts a client content request in an IP network, and transparently directs the content request to a best-fit server. The best-fit server is chosen based on the type of content requested, the quality of service requirements implied by the content request, the degree of load on available servers, network congestion information, and the proximity of the client to available servers. The flow switch detects client-server flows based on the arrival of TCP SYNs and/or HTTP GETs from the client. The flow switch implicitly deduces the quality of service requirements of a flow based on the content of the flow. The flow switch also provides the functionality of multiple physical web servers on a single web server in a way that is transparent to the client, through the use of virtual web hosts and flow pipes.

Claims

1. In a network, a method for directing packets between a client and a server, the method comprising:

receiving a client request for content via the network;

deriving, from the client request, content information descriptive of a plurality of characteristics of the content requested by the client request;

in response to receiving the client request, selecting a server from among a set of candidate servers based on

i) the derived content information; and

ii) a combination of server metrics obtained after receipt of the client request from all available servers capable of servicing the client request for content;

subsequently forwarding to the selected server transmissions originating from the client which are associated with the client request for content; and

subsequently forwarding to the client transmissions originating from the selected server which are associated with the client request for content.

2. The method of claim 1, wherein the client request is an HTTP request.

3. The method of claim 2, wherein deriving content information comprises:

extracting information from at least one portion of an HTTP header of the client request.

4. The method of claim 2, wherein deriving content information comprises deriving content information based on a Universal Resource Locator (URL) included in the client request.

5. The method of claim 2, wherein deriving content information comprises deriving content information based on a filename included in the client request.

6. The method of claim 5, wherein deriving content information based on a filename comprises deriving content information based on the filename extension.

7. The method of claim 2, wherein deriving content information comprises deriving content information based on a port identified in the client request.

8. The method of claim 2, wherein deriving content information comprises deriving content information based on query parameters including in the client request.

9. The method of claim 8, wherein the query parameters comprise Common Gateway Interface (CGI) parameters included in a URL of the client request.

10. The method of claim 2, wherein the client request comprises one of the following:

an HTTP GET message, an HTTP HEAD message, an HTTP PUT message, and an HTTP POST message.

11. The method of claim 2, wherein deriving content information comprises extracting information from the body of the client request.

12. The method of claim 1, wherein the client request is a TCP request.

13. The method of claim 1, further comprising:

obtaining additional information from the client about the content requested by the client request; and

wherein the selecting further comprises selecting based on the additional information.

14. The method of claim 1, further comprising:

obtaining client capability information about the client; and

wherein the selecting further comprises selecting the selected server based on the client capability information.

15. The method of claim 1, wherein selecting as the server comprises:

determining whether the client request requires persistent connectivity with a particular candidate server;

if the client request requires persistent connectivity with a particular server, identifying a candidate server with which the client is persistently connected for service of the client request;

selecting the identified candidate server.

16. The method of claim 1, further comprising determining whether an active path exists between the client and the selected server.

17. The method of claim 16, wherein determining whether an active path exists comprises sending a PING packet to the client.

18. The method of claim 16, wherein determining whether an active path exists comprises performing a Border Gateway Protocol route table lookup.

19. The method of claim 16, wherein the location of the client comprises a continent in which the client resides.

20. The method of claim 19, wherein the locations of the plurality of servers are continents in which the servers reside.

21. The method of claim 16, wherein identifying servers that are in the same location as the client comprises:

identifying administrative authorities associated with the client;

identifying, for each of the plurality of servers, administrative authorities associated with the server; and

identifying servers associated with an administrative authority that is associated with the client.

22. The method of claim 21, wherein the administrative authorities are Internet Service Providers.

23. The method of claim 1, further comprising

deriving, from the client request, quality of service information descriptive of quality of service requirements of the content requested by the client request;

and

wherein the selecting further comprises selecting based on the quality of service information.

24. The method of claim 1, wherein the deriving quality of service information includes deriving quality of service information from the content information.

25. The method of claim 1, wherein the deriving quality of service information includes deriving quality of service information from a size of the content requested by the client request.

26. The method of claim 1, wherein quality of service requirements comprise a bandwidth.

27. The method of claim 1, wherein quality of service requirements comprise a delay.

28. The method of claim 1, wherein quality of service requirements comprise a frame loss ratio.

29. The method of claim 1, wherein deriving quality of service information comprises deriving quality of service information from the MIME content type of the client request.

30. The method of claim 1, wherein deriving information descriptive of the content comprises deriving information descriptive of the content type.

31. The method of claim 1, wherein the selecting further comprises selecting based on at least one server metric describing at least one expected level of service provided by at least one of the candidate servers when serving the requested content.

32. The method of claim 31, wherein the at least one server metric includes:

one or more metrics selected from the following group;

a metric descriptive of server availability, a metric descriptive of the current load of at least one of the candidate servers, a metric descriptive of recent activity on at least one of the candidate servers, a metric descriptive of network congestion between the client and at least one of the candidate servers, a metric descriptive of the number of active connections being maintained by at least one of the candidate servers, a metric descriptive of the response time of at least one of the candidate servers, information descriptive of one or more previous selections of candidate servers, and client-server proximity information descriptive of distances between the client and at least one of the candidate servers.

33. The method of claim 32, wherein client-server proximity information comprises information descriptive of a continent in which the client resides and a continent in which the server resides.

34. The method of claim 33, wherein client-server proximity information further comprises information descriptive of an administrative authority associated with the client and an administrative authority associated with the server.

35. The method of claim 34, wherein the administrative authorities are Internet Service Providers.

36. The method of claim 31, wherein the at least one server metric includes:

two or more metrics selected from the following group;

a metric descriptive of server availability, a metric descriptive of the current load of at least one of the candidate servers, a metric descriptive of recent activity on at least one of the candidate servers, a metric descriptive of network congestion between the client and at least one of the candidate servers, a metric descriptive of the number of active connections being maintained by at least one of the candidate servers, a metric descriptive of the response time of at least one of the candidate servers, information descriptive of one or more previous selections of candidate servers, and client-server proximity information descriptive of distances between the client and at least one of the candidate servers.

37. The method of claim 31, wherein the at least one server metric is obtained by querying a database.

38. The method of claim 31, wherein the at least one server metric is obtained by periodically querying servers in the Internet Protocol network.

39. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the candidate server is receiving a burst of requests for the content requested by the client request.

40. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether satisfying the client request will result in a short-term flow.

41. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the content requested by the client request has been frequently requested in the past.

42. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the content requested by the client request has a high priority.

43. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of a probability that the content requested by the client request is cached by the server.

44. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the candidate server has responded to recent queries.

45. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the candidate server recently responded to a request for the content requested by the client request with an indication that the content is not served by the candidate server.

46. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the candidate server is reachable.

47. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the candidate server'"'"'s cache resources are below a threshold level.

48. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the candidate server'"'"'s active network connections are below a threshold level.

49. The method of claim 31, wherein the expected level of service provided by a candidate server is descriptive of whether the candidate server'"'"'s network bandwidth is below a threshold level.

50. A system for directing a stream of packets between a client and a server, the system comprising:

a plurality of servers;

a switch coupled to the plurality of servers by an Internet Protocol network through one or more communication links, wherein the switch comprises;

means for receiving a client request for content via the Internet Protocol network;

means for deriving, from the client request, content information descriptive of a plurality of characteristics of the content requested by the content request;

means, responsive to the means for deriving, for selecting a server from among a set of candidate servers serving the content requested by the client request, based on the content information;

means for subsequently forwarding to the selected server transmissions originating from the client which are associated with the client request for content; and

means for subsequently forwarding to the client transmissions originating from the selected server which are associated with the client request for content.

51. The system of claim 50, wherein:

the candidate servers comprise HTTP servers.

52. A switch in an Internet Protocol network, comprising:

means for receiving a client request for content via the Internet Protocol network;

means for deriving, form the client request, content information descriptive of a plurality of characteristics of the content requested by the content request;

means, responsive to the means for deriving, for selecting a server from among a set of candidate servers serving the content requested by the client request, based on the content information;

means for subsequently forwarding to the selected server transmissions originating from the client which are associated with the client request for content; and

means for subsequently forwarding to the client transmissions originating from the selected server which are associated with the client request for content.

53. In an Internet Protocol network, a method for use in a network switch, the method directing packets between a client and a server, the method comprising:

receiving an HTTP (HyperText Transfer Protocol) client request for content via the Internet Protocol network at an ingress port of the switch;

determining the content requested by the client request based on a plurality of characteristics related to portions of the client request;

selecting a server from among a set of candidate servers based on the determining;

subsequently forwarding to the selected server packets originating from the client which are associated with the client request for content via a switch egress port; and

subsequently forwarding to the client transmissions originating from the selected server which are associated with the client request for content via a switch egress port.

54. The method of claim 53, wherein the determining comprises determining based on a URL (Universal Resource Locator) included in the request.

55. The method of claim 53, wherein the determining comprises determining based on the requested domain name included in the request.

56. The method of claim 53, wherein the determining comprises determining a type of content requested.