Network bottlenecks

• US 8,959,233 B2
• Filed: 02/23/2011
• Issued: 02/17/2015
• Est. Priority Date: 02/23/2011
• Status: Expired due to Fees

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Myrah, Michael G., Natrajan, Balaji
• Primary Examiner(s): COULTER, KENNETH R
• Application Number:
  US13/033,250
• Publication Number:
  US 20120215937A1
• Time in Patent Office: 1,455 Days
• Field of Search: 709227-229, 709/232, 709/235, 709/230
• US Class Current: 709/227
• 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 2201/81 Threshold

Abstract

A method comprises receiving a request for a network connection and determining if the requested network connection is available. Based on the network connection not being available, the method comprises incrementing a counter. Based on the counter exceeding a threshold value, the method comprises setting a status indicating a bottleneck condition and further responding to the status indicative of the bottleneck condition.

Claims

1. A method executing on computer hardware, comprising:

receiving a request for a network connection;

determining if the requested network connection is available;

based on the network connection not being available, incrementing a counter;

based on said counter exceeding a threshold value within a predetermined time period, setting a status indicating a network bottleneck condition;

determining a location of the network bottleneck condition from among a plurality of expanders downstream from one another; and

responding to said status indicative of the network bottleneck condition.

2. The method of claim 1, wherein receiving, determining, and incrementing are performed during the predetermined period measured by a timer.

3. The method of claim 2 further comprising resetting the timer and counter based on an expiration of the timer and repeating said receiving, determining, and incrementing after resetting said timer and counter.

4. The method of claim 1 further comprising recording a timestamp associated with setting the status.

5. The method of claim 1 further comprising generating a reply message indicating that the requested connection is in use and is being waited on.

6. The method of claim 1 further comprising, indicating which physical interface of a plurality of physical interfaces is associated with the bottleneck condition.

7. The method of claim 1 wherein responding to said status comprises at least one of generating a recommendation to add a cable, sending an email, generating a trap, and generate a warning.

8. The method of claim 1, further comprising:

determining a type of message received; and

forwarding a Connection Unavailable message to an upstream expander; and

incrementing a counter of a corresponding expander.

9. The method of claim 1, further comprising maintaining independent counters for each downstream PHY.

10. The method of claim 1, further comprising if a status of an upstream-most expander indicates a bottleneck, then continuing to read a status of a next subsequently downstream expander in a fabric.

11. A method executing on computer hardware, comprising:

reading status of a first device;

determining whether status of the first device is indicative of a bottleneck based on a counter exceeding a threshold value within a predetermined time period, said counter incremented when a network connection is not available;

based on determining that the status of the first device is indicative of a bottleneck, reading status of a second device that is downstream from the first device, and determining a location of the network bottleneck condition from among a plurality of expanders downstream from one another; and

based on determining that the status of the second device is not indicative of a bottleneck, determining that a congestion point occurred with respect to the second device.

12. The method of claim 11 further comprising based on determining that the status of the second device is indicative of the bottleneck, further reading a status of a third device, downstream from the second device, and determining whether the status of the third device is indicative of a bottleneck.

13. The method of claim 12 further comprising:

receiving, with the first device, a forwarded message from the second device, wherein said forwarded message indicates a request for a network connection is waiting to be granted;

forwarding said forwarded message to a device upstream from the first device; and

incrementing a counter based on said forwarded message.

14. The method of claim 13 further comprising repeating said receiving, forwarding, and incrementing, and further comprising determining whether the counter exceeds a threshold.

15. The method of claim 13 further comprising repeating said receiving, forwarding, and incrementing during a predetermined time period, and further comprising determining whether the counter exceeds a threshold during the predetermined time period.

16. The method of claim 15 further comprising resetting said counter after determining whether the counter exceeds a threshold.

17. The method of claim 15 further comprising setting the status of the first device to be indicative of a bottleneck based on the counter exceeding the threshold.

18. The method of claim 17 further comprising recording a timestamp associated with setting the status.

19. The method of claim 11 further comprising responding to the determination that a congestion point has occurred by way of a recommendation to add a cable.

20. A system, comprising:

a non-transitory computer-readable storage medium containing software; and

a processor coupled to the computer-readable storage medium and that executes the software;

wherein the software causes the processor to;

read a status of a first device;

determine whether the status of the first device is indicative of a network bottleneck based on a counter exceeding a threshold value within a predetermined time period, said counter incremented when a network connection is not available;

based on a determination that the status of the first device is indicative of a network bottleneck, read a status of a second device that is downstream from the first device, and determining a location of the network bottleneck condition from among a plurality of expanders downstream from one another; and

based on a determination that the status of the second device is not indicative of a bottleneck, determine that a congestion point occurred with respect to the second device.

21. The system of claim 20 further comprising, based on a determination that the status of the second device is indicative of the bottleneck, the software causes the processor to read a status of a third device, downstream from the second device, and determine whether the status of the third device is indicative of a bottleneck.

22. The system of claim 20 wherein the software further causes the processor to respond to the determination that a congestion point has occurred by way of at least one of providing a recommendation to add a cable, sending an email, generating an SNMP trap, and generating a warning.

23. The system of claim 20 wherein the first device comprises an upstream-most expander in a network fabric and said software causes the processor to cease reading the status of other devices for bottlenecks if the first device'"'"'s status indicates the absence of a bottleneck.