Switching system

• US 7,447,777 B1
• Filed: 02/11/2002
• Issued: 11/04/2008
• Est. Priority Date: 02/11/2002
• Status: Active Grant

• Current Assignee: Extreme Networks, Inc.
• Original Assignee: Extreme Networks, Inc.
• Inventors: Gomez, Miguel, de la Iglesia, Erik, Zhang, Hui, Chou, Chien C., Wayne, Todd E., Carrie, Susan, Wang, Wei, Singh Ahuja, Ratinder Paul, Wallace, Mark Aaron, TRAN, KIET, Puri, Rahoul, Lowe, Ricky K., Liu, Liang
• Primary Examiner(s): Burgess, Glenton B.
• Assistant Examiner(s): Nash, Lashanya R
• Application Number:
  US10/098,957
• Time in Patent Office: 2,458 Days
• Field of Search: 709227-228, 709/229, 709/226
• US Class Current: 709/227
• CPC Class Codes: H04L 12/66 Arrangements for connecting between networks having differing types of switching systems, e.g. gateways
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

Systems and related methods are described for handling one or more resource requests. A protocol engine receives a resource request in accordance with a prescribed protocol, and a classification engine determines a desired class of service for the request. An analysis engine optionally analyzes the request, and, responsive thereto, determines a desired sub-class of service for the request. A policy engine then allocates a resource to the request responsive to one or both of the desired class of service, and the desired sub-class of service.

Claims

1. A system for handling one or more resource requests comprising:

a protocol engine for receiving a resource request in accordance with a prescribed protocol;

a classification engine for determining a desired class of service for the resource request;

an interface engine for checking the resource request for errors, wherein the classification engine and the interface engine concurrently evaluate the resource request;

an analysis engine for determining a desired sub-class of service for and responsive to the resource request; and

a policy engine for allocating a resource to the request responsive to one or both of the desired class of service, and the desired sub-class of service;

wherein the one or more resource requests are in the form of packets;

wherein the interface engine checks a packet for errors;

wherein the interface engine, if it detects an error in a packet, flushes the packet from the system before the packet is sent to its desired destination;

wherein the classification engine derives one or more keys from a packet, and uses the same to determine the desired class of service for the packet; and

wherein the classification engine uses a first key derived from the packet to access a database to determine if a connection already exists for the packet, and, if so, arranges to have the protocol engine forward the packet to the destination called for by the connection.

2. The system of claim 1 wherein the classification engine uses a second key derived from a packet to determine a service index for the packet.

3. The system of claim 1 wherein the resource allocated by the policy engine is a server.

4. The system of claim 3 wherein the protocol engine binds the server to an entity to create a connection.

5. The system of claim 1 wherein a resource allocated to the request is represented by a connection, which in turn is represented by a pointer having a 1-1 relationship with the connection, and pointing to a control block for the connection.

6. The system of claim 5 wherein a connection can be a client-system connection, a system-server connection, or a client-server connection.

7. The system of claim 5 wherein the control block having a 1-1 relationship with the connection can represent a client-system connection, a system-server connection, or a client-server connection.

8. The system of claim 5 wherein a connection, when created, is assigned a generation count, and a server is bound to the connection only if the generation count for the connection equals an expected value.

9. The system of claim 1 wherein a packet has a packet header that specifies the system as the destination of the packet, and the protocol engine modifies the packet header to specify a client or server as the destination of the packet.

10. The system of claim 1 wherein a packet has a packet header that specifies the system as the destination of the packet, and the protocol engine modifies the packet header to specify a client or server as the destination of the packet;

and

wherein the protocol engine determines the client or server as the destination for the packet from one or more control blocks for a client-server connection that exists for the packet.

11. The system of claim 1 wherein, for a first category of service, the system functions as a proxy for a server in relation to the packet, and, for a second category of service, does not function as a proxy for a server in relation to the packet.

12. The system of claim 1 wherein, for a selected category of service, the system buffers packets received from clients for assembly into larger data groupings, but does not do so in relation to packets received from servers for the selected category of service.

13. The system of claim 12 wherein, for a selected category of service, the system checks for contiguity in the assembly buffer for packets received from clients, but not for packets received from servers for the selected category of service.

14. The system of claim 1 wherein, for a first category of service, the system allocates a server to a packet responsive to content of a data grouping encapsulating the packet, while, for a second category of service, the system allocates a server to a packet responsive to information other than content of a data grouping encapsulating the packet.

15. The system of claim 1 wherein, for a first category of service, the system allocates a server to the packet after analyzing content of a data grouping encapsulating the packet to determine a desired sub-class of service for the packet, while, for a second category of service, the system allocates a server to the packet upon ascertaining the desired class of service for the packet.

16. The system of claim 1 wherein, for a selected category of service, the system acknowledges packets received from clients, but does not do so in relation to packets received from servers for the selected category of service.

17. The system of claim 1 wherein each of the engines are implemented in hardware, or a combination of hardware and software.

18. The system of claim 17 wherein each of the engines are implemented as one or more finite state machines.

19. The system of claim 1 wherein the desired class of service comprises a class of service at a layer in an applicable reference model corresponding to OSI layer four or higher.

20. The system of claim 19 wherein the desired class of service comprises a class of service at a layer in an applicable reference model corresponding to OSI layer six or higher.

21. The system of claim 1 wherein the desired sub-classes of service is based on content.

22. The system of any of claim 11, 14, or 15, wherein the first category of service comprises an OSI layer 5 class of service, and the second category of service comprises an OSI layer 4 class of service.

23. The system of any of claim 12, 13, or 16, wherein the selected category of service comprises an OSI layer 5 category of service.

24. A method for handling one or more resource requests comprising the steps of:

receiving a resource request in accordance with a prescribed protocol;

determining a desired class of service for the request;

checking the resource request for errors while determining the desired class of service for the request;

analyzing the request, and, responsive thereto, determining a desired sub-class of service for the request; and

allocating a resource to the request responsive to one or both of the desired class of service and the desired sub-class of service;

wherein a resource request is in the form of a packet, and the method further comprises

checking a packet for errors, and, if an error is detected in the packet, flushing the packet before sending the packet to its desired destination;

determining the desired class of service for a packet through one or more keys derived from the packet; and

using a first key derived from the packet to access a database ad determine therefrom if a connection for the packet already exists and, if so, forwarding the packet to a destination called for by the connection.

25. The method of claim 24 wherein the resource is a server.

26. The method of claim 25 further comprising binding the server to an entity to give rise to a connection.

27. The method of claim 24 further comprising representing a connection to the resource by a unique pointer which points to a control block for the connection.

28. The method of claim 27 wherein a packet has a packet header, and the method further comprises modifying the packet header to specify a client or server as the destination of the packet.

29. The method of claim 24 wherein a packet has a packet header, and the method further comprises:

modifying the packet header to specify a client or server as the destination of the packet; and

determining the client or server as the destination for the packet from one or more control blocks for a client-server connection which exists for the packet.

30. The method of claim 24 further comprising, for a first category of service, functioning as a proxy for a server in relation to the packet, and, for a second category of service, not functioning as a proxy for a server in relation to the packet.

31. The method of claim 29 further comprising, for a selected category of service, buffering packets received from clients for assembly into larger data groupings, but not doing so in relation to packets received from servers for the selected category of service.

32. The method of claim 31 further comprising, for a selected category of service, acknowledging packets received from clients, but not from servers.

33. The method of claim 24 further comprising, for a selected category of service, placing packets received from clients in an assembly buffer, and checking for contiguity in the assembly buffer, but not doing so for packets received from servers for the selected category of service.

34. The method of claim 24 wherein the resource is a server and

the method further comprises:

binding the server to an entity to give rise to a connection; and

assigning to a connection a generation count when the connection is created, and binding a server to the connection only if the generation count for the connection equals an expected value.

35. The method of claim 24 further comprising, for a first category of service, allocating a server to the packet after analyzing the content of a data grouping encapsulating the packet, and, for a second category of service, allocating a server to the packet upon ascertaining the desired class of service for the packet.

36. The method of claim 24 wherein the desired sub-class of service is based on content.

37. The method of claim 29 further comprising, for a first category of service, allocating a server to a packet responsive to the content of a data grouping encapsulating the packet, and, for a second category of service, allocating a server to a packet responsive to information other than the content of a data grouping encapsulating the packet.

38. The method of claim 29 wherein the desired class of service is selected from the group comprising an OSI layer 4 class of service, and an OSI layer 5 class of service.

39. The method of claim 38 wherein group further comprises an OSI layer 6 class of service and an OSI layer 7 class of service.

40. The method of any of claim 30, 37, or 35, wherein the first category of service comprises an OSI layer 5 class of service, and the second category of service comprises an OSI layer 4 class of service.

41. The method of any of claim 31 or 32 wherein the selected category of service comprises an OSI layer 5 class of service.