Transparent recovery of switch device

• US 20060059269A1
• Filed: 09/13/2004
• Published: 03/16/2006
• Est. Priority Date: 09/13/2004
• Status: Abandoned Application

• Current Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Original Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Inventors: Lyu, Allen, Lih, Yolin, Schober, Richard L., Talaat, Mohamed Magdy, Colloff, Ian, Chen, Chien, Reeve, Richard J.
• Application Number:
  US10/939,531
• Publication Number:
  US 20060059269A1
• Time in Patent Office: Days
• US Class Current: 709/235
• CPC Class Codes: H04L 49/254 Centralised controller, i.e. arbitration or scheduling
• CPC Class Codes: H04L 49/55 Prevention, detection or correction of errors
• CPC Class Codes: H04L 49/555 Error detection
• CPC Class Codes: H04L 49/557 Error correction, e.g. fault recovery or fault tolerance

Abstract

An interconnect device for transmitting data packets includes a plurality of ports, a hub, and an arbiter. The hub is configured to connect the plurality of ports together. The arbiter is coupled to the hub for controlling transmission of data packets between the hub and the ports. A reset is provided in at least one of the ports. The reset is in communication with the arbiter such that arbiter can reset the port in response to a detected error in the port.

Claims

1. An interconnect device for transmitting data packets, the interconnect device comprising:

a plurality of ports;

a hub connecting the plurality of ports;

an arbiter coupled to the hub for controlling transmission of data packets between the hub and the ports; and

a reset block in at least one of the ports and in the arbiter, the reset blocks in communication with the arbiter such that arbiter can initiate the reset block to reboot the at least one port in response to a detected error in the at least one port.

2. The interconnect device of claim 1, wherein the arbiter resets the at least one port without any software intervention from outside the interconnect device.

3. The interconnect device of claim 1, wherein the arbiter resets the at least one port without loosing configuration information for the interconnect device.

4. The interconnect device of claim 1, wherein each of the plurality of ports further comprise a physical block and a link block.

5. The interconnect device of claim 4, wherein the reset block is contained in the physical block and the link block of each of the plurality of ports.

6. The interconnect device of claim 1, wherein the arbiter is configured to negotiate with partners that are in communication with the interconnect device when the interconnect device is in a reboot such that packets that are flushed with the reboot may be retransmitted to the partner.

7. The interconnect device of claim 1, wherein the arbiter reboots only the port that is affected by the error and does not reboot any ports that are not affected by the error.

8. The interconnect device of claim 1, wherein the arbiter reboots all of the ports in the interconnect device when any of the ports are affected by the error.

9. The interconnect device of claim 1, wherein interconnect device logs the reboot of the switch such that the interconnect device can be subsequently polled with regard to the error and the reboot.

10. The interconnect device of claim 1, wherein arbiter reboots the switch in less than 100 microseconds.

11. The interconnect device of claim 1, wherein the interconnect device is in InfiniBand switch.

12. An InfiniBand network comprising:

a plurality of end nodes;

a subnet manager in communication with the end nodes; and

at least one InfiniBand switch, the at least one InfiniBand switch further comprising;

a plurality of ports;

a hub connecting the plurality of ports;

an arbiter coupled to the hub for controlling transmission of data packets between the hub and the ports; and

a reset block in at least one of the ports and in the arbiter, the reset blocks in communication with the arbiter such that arbiter can reboot the at least one port in response to a detected error in the at least one port.

13. The InfiniBand network of claim 12, wherein the arbiter resets the at least one port in the InfiniBand switch without any intervention from the subnet manager.

14. The InfiniBand network of claim 12, wherein the arbiter in the InfiniBand switch resets the at least one port without loosing configuration information for the InfiniBand switch.

15. The InfiniBand network of claim 12, wherein each of the plurality of ports in the InfiniBand switch further comprise a physical block and a link block wherein the reset block is contained in the physical block and in the link block of each of the plurality of ports.

16. The InfiniBand network of claim 12, wherein the arbiter in the InfiniBand switch reboots only the port that is affected by the error and does not reboot any ports that are not affected by the error.

17. The InfiniBand network of claim 12, wherein the arbiter in the InfiniBand switch reboots all of the ports in the InfiniBand switch when any of the ports are affected by the error.

18. A method for rebooting an interconnect device with an arbiter and with a plurality of ports connected by a hub that is configured to transmit data packets, the method comprising:

detecting when an error occurs in a port in the transmission of data packets in the plurality of ports;

placing the port affected by the error in an active deferred state;

initiating the rebooting of the interconnect device by the arbiter within the interconnect device resetting the port affected by the error; and

wherein the rebooting of the interconnect device occurs without any software intervention from outside the interconnect device.

19. The method of claim 18 wherein the arbiter resets the port affected by the error without loosing configuration information for the interconnect device.

20. The method of claim 18 wherein the arbiter negotiates with partners that are in communication with the interconnect device when the interconnect device is in a reboot such that packets that are flushed with the reboot may be retransmitted to the partner.