Packet forwarding optimization without an intervening load balancing node

• US 9,882,972 B2
• Filed: 10/30/2015
• Issued: 01/30/2018
• Est. Priority Date: 10/30/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Herr, David A., Mericle, Grant S., Ratcliff, Bruce H., Rau, Benjamin T.
• Primary Examiner(s): CHAN, SAI MING
• Application Number:
  US14/928,231
• Publication Number:
  US 20170126791A1
• Time in Patent Office: 823 Days
• Field of Search: 370389
• CPC Class Codes: H04L 45/74 Address processing for routing
• CPC Class Codes: H04L 47/50 Queue scheduling
• CPC Class Codes: H04L 61/5007 Internet protocol [IP] addresses
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers

Abstract

According to some embodiments, a computer-implemented method for packet forwarding optimization across multiple nodes may include receiving, via a processor on a network controller, a first packet from a client. The first packet can include information indicative of a target node and a protocol type. The method may further include receiving, via the processor, a balancing node from a packet engine, forwarding, via the processor, the first packet to the target node though the balancing node according to a balancing node address, a target node address, and the protocol type, receiving, via the processor, a registration from the packet engine, and saving the registration to a network controller queue, and forwarding, via the network controller queue, a second packet to the target node based on the registration without the second packet traversing the balancing node.

Claims

1. A computer-implemented method for packet forwarding optimization across multiple nodes utilizing a single internet protocol (IP) address, comprising:

receiving, via a processor on a network controller, a first packet from a client, wherein the first packet includes information indicative of a target node and a protocol type;

receiving, via the processor, a balancing node address of a balancing node from a packet engine;

forwarding, via the processor, the first packet to the target node though the balancing node according to the balancing node address, a target node address, and the protocol type;

receiving, via the processor, a registration from the packet engine, and saving the registration to a network controller queue, wherein the registration comprises the target node address mapped by the packet engine, wherein the registration does not include the balancing node address; and

forwarding, via the network controller queue, a second packet to the target node based on the registration without the second packet traversing the balancing node.

2. The computer-implemented method of claim 1, wherein the target node and the balancing node utilize the single IP address, and the client has an IP address different from the single IP address.

3. The computer-implemented method of claim 1, wherein the registration comprises an IP protocol type, an originating node address, the target node address, an originating node port number, and a target node port number.

4. The computer-implemented method of claim 3, wherein the registration comprises a plurality of target node addresses.

5. The computer-implemented method of claim 1, wherein the originating node and the target node are shared on the network controller.

6. The computer-implemented method of claim 1, wherein the forwarding the first packet to the target node is based on the target node address and the protocol type.

7. A system for packet forwarding optimization across multiple nodes utilizing a single internet protocol (IP) address, comprising:

a network controller comprising a processor configured to;

receive a first packet from a client, wherein the first packet includes information indicative of a target node address and a protocol type;

receive a balancing node address of a balancing node from a packet engine;

forward first packet to the target node though the balancing node according to the balancing node address, a target node address, and the protocol type;

receive a registration from the packet engine and save the registration to a network controller queue, wherein the registration comprises the target node address mapped by the packet engine, and wherein the registration does not include the balancing node address; and

forward, via the network controller queue, a second packet to the target node based on the registration without the second packet traversing the balancing node.

8. The system of claim 7, wherein the target node and the balancing node utilize the single IP address, and the client has an IP address different from the single IP address.

9. The system of claim 7, wherein the registration comprises an IP protocol type, an originating node address, the target node address, and an originating node port number, and target node port number.

10. The system of claim 9, wherein the registration comprises a plurality of target node addresses.

11. The system of claim 7, wherein the originating node and the target node are shared on the network controller.

12. The system of claim 7, wherein the forwarding the first packet to the target node is based on the shared destination ID and the protocol type.

13. A non-transitory computer-readable storage medium storing executable instructions that, when executed by a computer, perform a method for packet forwarding optimization across multiple nodes utilizing a single internet protocol (IP) address, the method comprising:

receiving, via a processor on a network controller, a first packet from a client, wherein the first packet includes information indicative of a target node address and a protocol type;

receiving, via the processor, a balancing node address of a balancing node from a packet engine;

forwarding, via the processor, the first packet to the target node though the balancing node according to the balancing node address, a target node address, and the protocol type;

receiving, via the processor, a registration from the originating node, and saving the registration to a network controller queue, wherein the registration comprises the target node address mapped by the packet engine, wherein the registration does not include the balancing node address; and

forwarding, via the network controller queue, a second packet to the target node based on the registration without the second packet traversing the balancing node.

14. The non-transitory computer-readable storage medium of claim 13, wherein the target node and the balancing node utilize the single IP address, and the client has an IP address different from the single IP address.

15. The non-transitory computer-readable storage medium of claim 13, wherein the registration comprises an IP protocol type, an originating node address, the target node address, and an originating node port number, and a target node port number.

16. The non-transitory computer-readable storage medium of claim 15, wherein the registration comprises a plurality of target node addresses.

17. The non-transitory computer-readable storage medium of claim 13, wherein the originating node and the target node are shared on the network controller.

18. The non-transitory computer-readable storage medium of claim 13, wherein the forwarding the first packet to the target node is based on the target node address and the protocol type.