Network proxy for high-performance, low-power data center interconnect fabric

• US 9,311,269 B2
• Filed: 12/03/2012
• Issued: 04/12/2016
• Est. Priority Date: 10/30/2009
• Status: Active Grant

• Current Assignee: Calxeda Inc.
• Original Assignee: III Holdings 2, LLC
• Inventors: Borland, David James, Goss, Kenneth S., Davis, Mark B, Volpe, Thomas A
• Primary Examiner(s): Rinehart, Mark
• Assistant Examiner(s): Hunt, Kenneth P
• Application Number:
  US13/692,741
• Publication Number:
  US 20140359089A1
• Time in Patent Office: 1,226 Days
• Field of Search: 370351-431, 709212-219, 709238-245, 710 22- 28, 710260-269, 711100-105, 711/108, 711147-153, 711170-200, 713300-340
• US Class Current: 1/1
• CPC Class Codes: G06F 1/3209 Monitoring remote activity, e.g. over telephone lines or network connections
• CPC Class Codes: G06F 1/3234 Power saving characterised by the action undertaken
• CPC Class Codes: G06F 1/3287 by switching off individual functional units in the computer system
• CPC Class Codes: G06F 1/3293 by switching to a less power-consuming processor, e.g. sub-CPU
• CPC Class Codes: G06F 13/00 Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units interface circuits for specific input/output devices G06F3/00 ; multiprogram control therefor G06F9/46; multiprocessor systems G06F15/16
• CPC Class Codes: G06F 13/24 using interrupt G06F13/32 takes precedence
• CPC Class Codes: G06F 13/40 Bus structure for computer networks G06F15/163; for optical bus networks H04B10/25
• CPC Class Codes: G06F 13/4022 using switching circuits, e.g. switching matrix, connection or expansion network G06F13/4009 takes precedence
• CPC Class Codes: G06F 15/177 Initialisation or configuration control processor initialisation G06F9/4405
• CPC Class Codes: H04L 45/00 Routing or path finding of packets in data switching networks routing or path finding in wireless networks H04W40/00
• CPC Class Codes: H04L 45/60 Router architectures
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 49/109 Integrated on microchip, e.g. switch-on-chip
• CPC Class Codes: H04L 49/25 Routing or path finding in a switch fabric
• CPC Class Codes: H04L 49/3009 Header conversion, routing tables or routing tags
• CPC Class Codes: H04L 49/351 for local area network [LAN], e.g. Ethernet switches
• CPC Class Codes: H04L 49/356 for storage area networks
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management
• CPC Class Codes: Y02D 30/00 Reducing energy consumption in communication networks
• CPC Class Codes: Y02D 30/50 in wire-line communication networks, e.g. low power modes or reduced link rate

Abstract

A system and method are provided for network proxying. The network proxying may occur in a node of a fabric or across nodes in the fabric. In the network proxying, the node has a processor with a low power mode and the system remaps, by a management processor of the node, a port identifier for a processor that is in a low power mode to the management processor. The management processor then processes a plurality of packets that contain the port identifier for the processor that is in the low power mode to maintain a network presence of the node.

Claims

1. A node, comprising:

at least one processor associated with at least one first internal port, wherein the at least one first internal port has a first address;

a management processor associated with a second internal port, wherein the second internal port has a second address; and

a port remapping component configured to indicate that the first address is remapped to the second address, wherein the port remapping component is configured to activate a port remapping enable field, and wherein the port remapping enable field is specific to a given channel;

wherein the management processor is configured to disable a first-in first-out buffer of the at least one processor and to receive traffic destined for the at least one processor as a proxy of the at least one processor via the port remapping component and in response to the at least one processor being in a low power state.

2. The node of claim 1, wherein the management processor is further configured to process an incoming packet that is to be remapped to the management processor.

3. The node of claim 2, wherein the management processor is further configured to respond to the incoming packet if required for a particular protocol.

4. The node of claim 2, wherein the management processor is further configured to ignore the incoming packet if destined for another node.

5. The node of claim 2, wherein the management processor is further configured to wake up the at least one processor, if in the low power state, for the incoming packet that is to be processed by the at least one processor.

6. The node of claim 2, wherein the incoming packet is a keep alive message, and wherein the management processor is further configured to respond to the keep alive message.

7. The node of claim 3, wherein the incoming packet required for a protocol is an ARP packet, a NetBIOS datagram, or an ICMP ping.

8. The node of claim 4, wherein the incoming packet destined for another node is a broadcast packet or a unicast packet.

9. The node of claim 5, wherein the incoming packet that is to be processed by the at least one processor is a TCP SYN packet or a magic packet.

10. The node of claim 5, wherein the management processor is further configured to:

reset the port remapping component to cancel the remapping; and

resend the incoming packet back to the at least one processor.

11. A method, comprising:

associating at least one processor in a node with at least one first internal port, wherein the at least one first internal port has a first address;

associating a management processor in the node with a second internal port, wherein the second port has a second address;

setting up a port remapping component which indicates that the first address is remapped to the second address, wherein the port remapping component is configured to activate a port remapping enable field, and wherein the port remapping enable field is specific to a given channel;

disabling, by the management processor, a first-in first-out buffer of the at least one processor; and

receiving, by the management processor as a proxy of the at least one processor, traffic destined for the at least one processor via the port remapping component and in response to the at least one processor being in a low power state.

12. The method of claim 11, further comprising responding, by the management processor, to a packet remapped to the management processor if required for a particular protocol.

13. The method of claim 11, further comprising ignoring, by the management processor, a packet remapped to the management processor if destined for another node.

14. The method of claim 11, further comprising waking up the at least one processor, if in the low power state, for a packet that is to be processed by the at least one processor.

15. The method of claim 11, further comprising responding, by the management processor, to a keep alive message that is remapped to the management processor.

16. The method of claim 12, wherein the packet is an ARP packet, a NetBIOS datagram, or an ICMP ping.

17. The method of claim 13, wherein the packet destined for another node is a broadcast packet or a unicast packet.

18. The method of claim 14, wherein the packet that is to be processed by the at least one processor is a TCP SYN packet or a magic packet.

19. The method of claim 14, further comprising:

resetting the port remapping component to cancel the remapping; and

resending a packet back to the at least one processor.

20. A fabric, comprising:

a plurality of nodes interconnected to each other, wherein each of the plurality of nodes includes;

at least one processor, wherein the at least one processor is associated with at least one internal port of a node in which the at least one processor is located, and wherein the at least one internal port has a first address in the fabric;

a management processor, wherein the management processor is associated with a second internal port of the node, and wherein the second internal port has a second address in the fabric; and

a port remapping component configured to indicate that the first address is remapped to the second address, wherein the port remapping component is configured to activate a port remapping enable field, and wherein the port remapping enable field is specific to a given channel;

wherein the management processor is configured to disable a first-in first-out buffer of the at least one processor and to receive traffic destined for the at least one processor as a proxy of the at least one processor via the port remapping component and in response to the at least one processor being in a low power state.

21. The fabric of claim 20, wherein the management processor is further configured to process an incoming packet that is to be remapped to the management processor.

22. The fabric of claim 21, wherein the management processor is further configured to respond to the incoming packet if required for a particular protocol.

23. The fabric of claim 21, wherein the management processor is further configured to ignore the incoming packet if destined for another node.

24. The fabric of claim 21, wherein the management processor is further configured to wake up the at least one processor, if in the low power state, for the incoming packet that is to be processed by the at least one processor.

25. The fabric of claim 21, wherein the incoming packet is a keep alive message, and wherein the management processor is further configured to respond to the keep alive message.

26. The fabric of claim 22, wherein the incoming packet for a protocol is an ARP packet, a NetBIOS datagram, or an ICMP ping.

27. The fabric of claim 23, wherein the incoming packet destined for another node is a broadcast packet or a unicast packet.

28. The fabric of claim 24, wherein the incoming packet that is to be processed by the at least one processor is a TCP SYN packet or a magic packet.

29. The fabric of claim 24, wherein the management processor is configured to:

reset the port remapping component to cancel the remapping; and

resend the incoming packet back to the at least one processor.

30. An apparatus, comprising:

a plurality of nodes interconnected to each other to form a switching fabric, wherein each of the plurality of nodes is associated with a respective node ID in the switching fabric;

a proxy node configured to be a proxy for at least one of the plurality of nodes that have entered a low power state; and

a node proxy component configured to indicate that at least one of the respective node IDs is proxied to the proxy node, wherein the node proxy component comprises a routing table proxy component configured to determine if a node ID is being proxied to the proxy node;

wherein the proxy node is further configured to receive traffic destined for at least one of the plurality of nodes via the node proxy component and in response to the at least one of the plurality of nodes being in a low power state.

31. The apparatus of claim 30, wherein the routing table proxy component comprises a routing table with a proxy field configured to determine if a node ID is being proxied to the proxy node.

32. The apparatus of claim 30, wherein the proxy node is further configured to respond to an incoming packet, remapped to the proxy node, if required for a particular protocol.

33. The apparatus of claim 30, wherein the proxy node is further configured to ignore an incoming packet, remapped to the proxy node, if destined for a node that is not being proxied by the proxy node.

34. The apparatus of claim 30, wherein the proxy node is further configured to wake up a proxied node for an incoming packet that is to be processed by the proxied node.

35. The apparatus of claim 30, wherein the proxy node is further configured to respond to a keep alive message remapped to the proxy node.

36. The node of claim 32, wherein the incoming packet for a protocol is an ARP packet, a NetBIOS datagram, or an ICMP ping.

37. The node of claim 33, wherein the incoming packet destined for another node is a broadcast packet or a unicast packet.

38. The node of claim 34, wherein the incoming packet that is to be processed by the proxied node is a TCP SYN packet or a magic packet.

39. The apparatus of claim 34, wherein the proxy node is configured to:

reset the node proxy component to cancel the proxy; and

resend the incoming packet back to the proxied node.

40. A method, comprising:

associating each of a plurality of nodes with a respective node ID, wherein the plurality of nodes are interconnected to each other to form a switching fabric;

setting up a node proxy component which indicates that at least one of the respective node IDs is proxied to a proxy node, wherein the proxy node is configured to be a proxy for at least one of the plurality of nodes in response to the at least one of the plurality of nodes being in a low power state, wherein the node proxy component comprises a routing table proxy component configured to determine if a node ID is being proxied to the proxy node; and

receiving, by the proxy node, traffic destined for the at least one of the plurality of nodes via the node proxy component and in response to the at least one of the plurality of nodes being in a low power state.

41. The method of claim 40, wherein the node proxy component further comprises a routing table with a proxy field that determines if a node ID is being proxied to the proxy node.

42. The method of claim 40, wherein the proxy node is further configured to respond to an incoming packet, remapped to the proxy node, if required for a particular protocol.

43. The method of claim 40, wherein the proxy node is further configured to ignore an incoming packet, remapped to the proxy node, if destined for a node that is not being proxied by the proxy node.

44. The method of claim 40, wherein the proxy node is further configured to wake up a proxied node for an incoming packet that is to be processed by the proxied node.

45. The method of claim 40, wherein the proxy node is further configured to respond to a keep alive message.

46. The method of claim 42, wherein the incoming packet is an ARP packet, a NetBIOS datagram, or an ICMP ping.

47. The method of claim 43, wherein the incoming packet destined for another node is a broadcast packet or a unicast packet.

48. The method of claim 44, wherein the incoming packet that is to be processed by the proxied node is a TCP SYN packet or a magic packet.

49. The method of claim 44, wherein the proxy node is further configured to:

reset the proxy node component to cancel the proxy; and

resend the incoming packet back to the proxied node.