Load-balancing utilizing one or more threads of execution for implementing a protocol stack

• US 20040174877A1
• Filed: 03/03/2003
• Published: 09/09/2004
• Est. Priority Date: 03/03/2003
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Poon, Kacheong, Masputra, Cahya Adi
• Granted Patent: US 7,400,581 B2
• Time in Patent Office: Days
• US Class Current: 370/392
• CPC Class Codes: H04L 49/90 Buffering arrangements
• CPC Class Codes: H04L 49/901 using storage descriptor, e.g. read or write pointers
• CPC Class Codes: H04L 49/9042 Separate storage for different parts of the packet, e.g. header and payload
• CPC Class Codes: H04L 49/9063 Intermediate storage in different physical parts of a node or terminal
• CPC Class Codes: H04L 49/9078 using an external memory or storage device
• CPC Class Codes: H04L 49/9094 Arrangements for simultaneous transmit and receive, e.g. simultaneous reading/writing from/to the storage element

Abstract

Embodiments of the present invention provide a novel method of implementing a network communication protocol. In one embodiment, the state of the network communication process is determined based upon one or more conditions. The protocol stack is then executed as a single thread of execution if the network communication process is in a first state. Alternatively, the protocol stack is executed as a receive thread of execution and a drainer thread of execution if the network communication process is in a second state.

Claims

1. A network device comprising:

a interface module communicatively coupled to a network for receiving a data packet;

a computer-readable medium for storing information and instructions;

a processor for acting upon said information and instructions to provide the functions comprising;

a protocol stack for processing a header of said data packet comprising;

a single thread of execution when a first state of said network device exists; and

a receive thread of execution and a drainer thread of execution when a second state of said network device exists; and

an application wherein a payload of said data packet is utilized.

2. The network device according to claim 1, wherein said protocol stack determines said first and second state from one or more conditions consisting of a system load, a trend of said system load, a traffic type, and a trend of said traffic type.

3. The network device according to claim 1, wherein said interface module comprises a network interface card.

4. The network device according to claim 1, wherein said protocol stack comprises a transmission control protocol/internet protocol (TCP/IP).

5. The network device according to claim 1, wherein said single thread of execution comprises:

a network layer;

an internet layer; and

a transport layer.

6. The network device according to claim 1, wherein said receive thread of execution comprises a network layer.

7. The network device according to claim 1, wherein said drainer thread of execution comprises a transport layer.

8. A method of performing network communication comprising:

buffering a plurality of data packets;

processing said plurality of data packets according to a protocol utilizing a single thread of execution during a first condition; and

processing said plurality of data packets according to a protocol utilizing a receive thread of execution and a drainer thread of execution during a second condition.

9. The method according to claim 8, wherein said first condition comprises a low system load.

10. The method according to claim 8, wherein said first condition comprises an urgent data packet type.

11. The method according to claim 8, wherein said second condition comprises a high system load.

12. The method according to claim 8, wherein said single thread of execution comprises;

receiving one of said plurality of data packets;

processing a header of said one of said plurality of data packets; and

delivering a payload of said one of said plurality of data packets to an application according to said header.

13. The method according to claim 8, wherein said receive thread of execution comprises receiving one of said plurality of data packet.

14. The method according to claim 13, wherein said drainer thread of execution comprises;

processing a header of said one of said plurality of data packets; and

delivering a payload of said one of said plurality of data packets to an application according to said header.

15. The method according to claim 8, wherein said single thread of execution comprises:

aggregating one or more of said plurality of data packets destined for an application;

splitting said aggregated one or more of said plurality of data packets to form an aggregated headers and an aggregated payloads;

processing said aggregated headers; and

delivering said aggregated payloads to said application according to said aggregated headers.

16. The method according to claim 8, wherein said receive thread of execution comprises:

aggregating one or more of said plurality of data packets destined for an application; and

splitting said aggregated one or more of said plurality of data packets to form an aggregated headers and an aggregated payloads.

17. The method according to claim 16, wherein said drainer thread of execution comprises:

processing said aggregated headers; and

delivering said aggregated payloads to said application according to sad aggregated headers.

18. A method of implementing a network communication protocol stack comprising:

determining a state of said network communication process;

executing a single thread of execution when said state comprises a low system load; and

executing a receive thread of execution and a drainer thread of execution when said state comprises a high system load.

19. The method according to claim 18, wherein said single thread of execution comprises:

receiving a data packet;

routing said data packet according to a header of said data packet;

controlling flow of said data packet according to said header; and

delivering a payload of said data packet to a destination application according to said header.

20. The method according to claim 18, wherein said single thread of execution comprises:

receiving a plurality of data packets;

aggregating said plurality of data packets;

splitting said aggregated data packets into an aggregate header and an aggregate payload;

creating metadata concerning said aggregate header and said aggregate payload;

routing said aggregate payload utilizing said aggregate header and said metadata;

controlling flow of said aggregate payload utilizing said aggregate header and said metadata; and

delivering said aggregate payload to a destination application according to said aggregate header and said metadata.

21. The method according to claim 18, wherein said receive thread of execution comprises:

receiving a data packet; and

routing said data packet according to a header of said data packet.

22. The method according to claim 18, wherein said drainer thread of execution comprises:

controlling flow of said data packet according to said header; and

delivering a payload of said data packet to a destination application according to said header.

23. The method according to claim 18, wherein said receive thread of execution comprises:

receiving a plurality of data packets;

aggregating said plurality of data packets;

splitting said aggregated data packets into an aggregate header and an aggregate payload; and

creating metadata concerning said aggregate header and said aggregate payload.

24. The method according to claim 18, wherein said drainer thread of execution comprises:

controlling flow of said aggregate payload utilizing said aggregate header and said metadata; and

delivering said aggregate payload to a destination application according to said aggregate header and said metadata.

25. A computer-readable medium having instructions which when executed by a network device implement a protocol stack comprising:

buffering a data packet;

determining a state of said network device;

processing a first and second portion of a header of said data packet utilizing a first thread of execution when said state comprises a low system load;

processing said first portion of said header of said packet utilizing a second thread of execution and handing-off processing of said second portion of said header to a third thread of execution, when said state comprises a high system load; and

delivering a payload of said data packet to a destination application.

26. The computer-readable medium according to claim 25, wherein said first thread of execution comprises a single thread of execution.

27. The computer-readable medium claim 26, wherein said single thread of execution comprises an internet protocol and a transmission control protocol.

28. The computer-readable medium according to claim 27, wherein said internet protocol performs said determining said state of said network device.

29. The computer-readable medium according to claim 27, wherein said transmission control protocol performs said determining said state of said network device.

30. The computer-readable medium according to claim 25, wherein said second thread of execution comprises a receive thread of execution.

31. The computer-readable medium according to claim 30, wherein said third thread of execution comprises a drainer thread of execution.

32. The computer-readable medium according to claim 31, wherein said receive thread of execution comprises an internet protocol.

33. The computer-readable medium according to claim 32, wherein said drainer thread of execution comprises a transmission control protocol.

34. The computer-readable medium according to claim 33, wherein said internet protocol performs said determining said state of said network device.

35. The computer-readable medium according to claim 33, wherein said transmission control protocol performs said determining said state of said network device.