METHOD AND APPARATUS FOR PROCESSING PROTOCOL MESSAGES FOR MULTIPLE PROTOCOL INSTANCES

• US 20100205612A1
• Filed: 02/10/2009
• Published: 08/12/2010
• Est. Priority Date: 02/10/2009
• Status: Active Grant

• Current Assignee: Piece Future Pte., Ltd.
• Original Assignee: Alcatel-Lucent SA (Nokia Corporation)
• Inventors: Bhatia, Jagjeet, Touve, Jeremy
• Granted Patent: US 8,589,593 B2
• Time in Patent Office: Days
• US Class Current: 719/314
• CPC Class Codes: H04L 69/18 Multiprotocol handlers, e.g. single devices capable of handling multiple protocols

Abstract

The invention includes a method and apparatus for processing protocol messages for multiple protocol instances. In one embodiment, a method for processing protocol messages includes receiving a plurality of messages for a plurality of processors where each received message is associated with one of the protocol instances, generating a processing request for each message, queuing the processing requests, and servicing the queues in a manner for arbitrating access by the queues to the processors for processing the messages. A processing request generated for a received message identifies one of the protocol instances with which the message is associated. The processing requests are queued using a plurality of queues associated with the respective plurality of protocol instances, where each processing request is queued in one of the queues associated with the one of the protocol instances with which the processing request is associated. The servicing of each queue includes reading a processing request from the queue, if the queue has at least one processing request queued therein, and causing the one of the processors with which the processing request is associated to process one of the messages associated with the protocol instance identified by the processing request. The queues may be serviced in a round-robin manner for arbitrating access by the queues to the processors, thereby enabling atomic processing of the messages.

Claims

1. A method for controlling processing of messages for a plurality of protocol instances, comprising:

receiving a plurality of messages for a plurality of processors, wherein each message is associated with one of the protocol instances;

generating, for each message, a processing request identifying the one of the protocol instances with which the message is associated;

queuing the processing requests using a respective plurality of queues associated with the protocol instances, wherein each processing request is queued in the queue associated with the one of the protocol instances with which the processing request is associated; and

servicing the queues in a manner for arbitrating access by the queues to the processors for processing the messages, wherein each of the queues having at least one processing request queued therein is serviced by reading a processing request from the queue and causing the one of the processors with which the retrieved processing request is associated to process one of the messages associated with the protocol instance identified by the processing request.

2. The method of claim 1, wherein receiving the plurality of messages at the plurality of processors comprises:

receiving the messages at a plurality of network interfaces, wherein each network interface is associated with one of the respective processors.

3. The method of claim 1, wherein each processor has a memory associated therewith, wherein each memory comprises a plurality of message queues associated with the respective protocol instances, wherein the method further comprises:

for each message;

queuing the message in the memory associated with the processor at which the message is received, wherein the message is queued in one of the message queues of the memory, wherein the one of the message queues is associated with the one of the protocol instances with which the message is associated.

4. The method of claim 1, wherein, for each message, generating the processing request for the message comprises:

determining the one of the protocol instances with which the message is associated;

determining a processor identifier of the processor for which the message is received; and

generating the processing request for the received message;

wherein the processing request comprises the processor identifier;

wherein the processing request is adapted for being directed to the one of the queues associated with the determined one of the protocol instances with which the message is associated.

5. The method of claim 1, wherein each processing request is propagated from the processor at which the associated message is received to an arbiter module maintaining the queues.

6. The method of claim 1, wherein, for each processing request, queuing the processing request comprises:

determining the one of the protocol instances with which the processing request is associated; and

writing the processing request to the one of the queues associated with the one of the protocol instances with which the processing request is associated.

7. The method of claim 1, wherein servicing the queues comprises:

selecting one of the queues;

determining a queue status associated with the selected queue; and

in response to a determination that the selected queue has a status of EMPTY or BUSY, selecting a next one of the queues to be serviced;

in response to a determination that the selected queue has a status of WAITING, reading one of the processing requests from the selected queue and servicing the processing request.

8. The method of claim 7, wherein servicing the processing request comprises:

determining the one of the protocol instances with which the processing request is associated;

determining one of the processors with which the processing request is associated; and

generating a processing notice for the processing request, wherein the processing notice identifies the one of the protocol instances with which the processing request is associated; and

propagating the processing notice to the one of the processors with which the processing request is associated.

9. The method of claim 8, further comprising:

receiving the processing notice at the one of the processors with which the processing request is associated;

determining, from the processing notice, the one of the protocol instances with which the processing request is associated; and

processing one of the messages of the one of the protocol instances with which the processing request is associated.

10. The method of claim 9, further comprising:

clearing the processing request from the one of the queues from which the processing request was read;

determining a state of the one of the queues from which the processing request was read;

updating a status of the one of the queues from which the processing request was read from BUSY to EMPTY or WAITING based on the state; and

selecting a next one of the queues to be serviced.

11. An apparatus for controlling processing of messages for a plurality of protocol instances, comprising:

a plurality of processors, each processor adapted for processing messages associated with the protocol instances, each processor adapted for generating a processing request for each message received by the processor where each processing request identifies one of the protocol instances with which the message is associated; and

an arbiter module in communication with said processors, the arbiter module comprising a respective plurality of queues associated with the protocol instances, the arbiter module adapted for;

queuing the processing requests in respective ones of the queues based on the protocol instances with which the respective processing requests are associated; and

servicing the queues in a manner for arbitrating access by the queues to the processors for processing the messages, wherein each queue having at least one processing request queued therein is serviced by reading a processing request from the queue and causing the one of the processors with which the retrieved processing request is associated to process one of the messages associated with the protocol instance identified by the processing request.

12. The apparatus of claim 11, further comprising:

a plurality of network interfaces associated with the respective plurality of processors.

13. The apparatus of claim 11, further comprising:

a plurality of memory modules associated with the plurality of processors, respectively;

wherein each memory module comprises a plurality of message queues associated with the respective protocol instances for storing messages according to the respective protocol instances of the messages.

14. The apparatus of claim 11, wherein each processor is adapted for generating a processing request for a received message by:

determining the one of the protocol instances with which the message is associated;

determining a processor identifier of the processor for which the message is received; and

generating the processing request for the received message;

wherein the processing request comprises the processor identifier;

wherein the processing request is adapted for being directed to the one of the queues associated with the determined one of the protocol instances with which the message is associated.

15. The apparatus of claim 11, wherein each processor is adapted for propagating processing requests to the arbiter module.

16. The apparatus of claim 11, wherein the arbiter module is adapted for queuing each processing request by:

determining the one of the protocol instances with which the processing request is associated; and

writing the processing request to the one of the queues associated with the one of the protocol instances with which the processing request is associated.

17. The apparatus of claim 11, wherein the arbiter module is adapted for servicing the queues by:

selecting one of the queues;

determining a queue status associated with the selected queue; and

in response to a determination that the selected queue has a status of EMPTY or BUSY, selecting a next one of the queues to be serviced;

or

in response to a determination that the selected queue has a status of WAITING, reading one of the processing requests from the selected queue and servicing the processing request.

18. The apparatus of claim 17, wherein the arbiter module is adapted for servicing the processing request by:

determining the one of the protocol instances with which the processing request is associated;

determining one of the processors with which the processing request is associated; and

generating a processing notice for the processing request, wherein the processing notice identifies the one of the protocol instances with which the processing request is associated; and

propagating the processing notice to the one of the processors with which the processing request is associated.

19. The apparatus of claim 11, wherein each processor is adapted for:

receiving a processing notice from the arbiter module, wherein the processing notice is generated by the arbiter module for a processing request read by the arbiter module from one of the queues selected by the arbiter module to be serviced;

determining, from the processing notice, one of the protocol instances with which the processing request is associated; and

processing one of the messages of the one of the protocol instances with which the processing request is associated.

20. The apparatus of claim 19, wherein the arbiter module is adapted for:

clearing the processing request from the one of the queues from which the processing request was read;

determining a state of the one of the queues from which the processing request was read;

updating a status of the one of the queues from which the processing request was read from BUSY to EMPTY or WAITING based on the state; and

selecting a next one of the queues to be serviced.

21. A computer readable storage medium storing a software program which, when executed by a computer, causes the computer to perform a method for controlling processing of messages for a plurality of protocol instances, the method comprising:

receiving a plurality of messages for a plurality of processors, wherein each message is associated with one of the protocol instances;

generating, for each message, a processing request identifying the one of the protocol instances with which the message is associated;

queuing the processing requests using a respective plurality of queues associated with the protocol instances, wherein each processing request is queued in the queue associated with the one of the protocol instances with which the processing request is associated; and

servicing the queues in a manner for arbitrating access by the queues to the processors for processing the messages, wherein each of the queues having at least one processing request queued therein is serviced by reading a processing request from the queue and causing the one of the processors with which the retrieved processing request is associated to process one of the messages associated with the protocol instance identified by the processing request.