METHOD AND SYSTEM FOR DETERMINING THE IMPACT OF FAILURES IN DATA CENTER NETWORKS

• US 20130232382A1
• Filed: 03/01/2012
• Published: 09/05/2013
• Est. Priority Date: 03/01/2012
• Status: Abandoned Application

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Jain, Navendu, Gill, Phillipa
• Application Number:
  US13/409,111
• Publication Number:
  US 20130232382A1
• Time in Patent Office: Days
• US Class Current: 714/48
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/2007 using redundant communication media
• CPC Class Codes: G06F 11/202 where processing functionality is redundant redundant communication control functionality G06F11/2005, redundant storage control functionality G06F11/2089
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: H04L 41/064 involving time analysis
• CPC Class Codes: H04L 41/069 using logs of notifications; Post-processing of notifications

Abstract

There is provided a method and system for determining an impact of failures in a data center network. The method includes identifying failures for the data center network based on data about the data center network and grouping the failures into failure event groups, wherein each failure event group includes related failures for a network element. The method also includes estimating the impact of the failures for each of the failure event groups by correlating the failures with traffic for the data center network.

Claims

1. A method for determining an impact of failures in a data center network, comprising:

identifying a plurality of failures for the data center network based on data about the data center network;

grouping the plurality of failures into a plurality of failure event groups, wherein each failure event group comprises a plurality of related failures for a network element; and

estimating the impact of the plurality of failures for each of the failure event groups by correlating the plurality of failures with traffic for the data center network.

2. The method of claim 1, wherein estimating the impact of the plurality of failures comprises:

computing a statistical measure of an amount of data transferred on a network link in a specified period of time;

computing a statistical measure of an amount of data transferred on the network link during the specified period of time; and

calculating a change in an amount of data that was transferred during the specified period of time based on the statistical measure.

3. The method of claim 2, wherein the specified period of time comprises a period of time preceding a failure, a period of time of the failure, or a period of time after the failure, or any combinations thereof.

4. The method of claim 2, wherein calculating the change in the amount of data comprises:

subtracting the statistical measure of the amount of data transferred on the network link during the period of the failure from the statistical measure of the amount of data transferred on the network link in the period preceding the failure to obtain a first value; and

multiplying the first value by a duration of the failure to obtain an estimate of the change in the amount of data that was transferred during the duration of the failure.

5. The method of claim 1, wherein estimating the impact of the plurality of failures comprises estimating an impact of a failure on a link by computing a ratio of a statistical measure of an amount of traffic on the link during the failure to a statistical measure of an amount of traffic on the link before the failure.

6. The method of claim 1, comprising determining an impact of a failure of a network device by applying the method of claim 1 across links and devices.

7. The method of claim 1, comprising estimating the impact of the plurality of failures based on a correlation across multiple data sources.

8. The method of claim 1, comprising:

determining an effectiveness of a network redundancy group of redundant network components comprising devices and links, in masking an impact of the plurality of failures for each of the plurality of failure event groups, by estimating a change in an amount of network traffic due to the plurality of failures by;

computing a statistical measure of an amount of data transferred on network links in a specified period of time preceding the failures;

computing a statistical measure of an amount of data transferred on the network links during the failures; and

calculating a change in an amount of data that was transferred during the failures based on a statistical measure across the network redundancy group.

9. The method of claim 8, wherein the statistical measure comprises a median.

10. A system for determining an impact of failures in a data center network, comprising:

a processor that is adapted to execute stored instructions; and

a system memory, wherein the system memory comprises code configured to;

identify a plurality of failures for the data center network based on data about the data center network;

group the plurality of failures into a plurality of failure event groups, wherein each failure event group comprises a plurality of related failures for a network element; and

estimate the impact of the plurality of failures for each of the plurality of failure event groups by correlating the plurality of failures with traffic for the data center network and data from multiple data sources.

11. The system of claim 10, wherein the system memory comprises code configured to determine an effectiveness of network redundancy groups in masking the impact of the plurality of failures for each of the plurality of failure event groups.

12. The system of claim 10, wherein the code configured to estimate the impact of the plurality of failures comprises code configured to:

compute a statistical measure of an amount of data transferred on a network link in a specified period of time;

compute a statistical measure of an amount of data transferred on the network link during the specified period; and

calculate a change in an amount of data that was transferred during the specified period based on the statistical measure.

13. The system of claim 10, wherein the impact of the plurality of failures comprises a change in an amount of network traffic due to the plurality of failures.

14. The system of claim 10, wherein estimating the impact of the plurality of failures comprises estimating an impact of a failure on a link by computing a ratio of a statistical measure of an amount of traffic on the link during the failure to a statistical measure of an amount of traffic on the link before the failure.

15. The system of claim 10, comprising estimating an effectiveness of network redundancy by computing a ratio of a statistical measure of an amount of traffic on links and devices within a network redundancy group during the failure to a statistical measure of an amount of traffic on the links and the devices within the network redundancy group before the failure.

16. One or more non-transitory, computer-readable storage media for storing computer-readable instructions, the computer-readable instructions providing a system for analyzing an impact of failures in a data center network when executed by one or more processing devices, the computer-readable instructions comprising code configured to:

identify a plurality of failures for the data center network based on data about the data center network, wherein the plurality of failures comprises one or more of a network device failure or a network link failure;

group the plurality of failures into a plurality of failure event groups, wherein each failure event group comprises a plurality of related failures for a network element;

determine the impact of the plurality of failures for each of the plurality of failure event groups by correlating the plurality of failures with a change in an amount of network traffic; and

determine an effectiveness of network redundancies in mitigating the impact of the plurality of failures for each of the plurality of failure event groups.

17. The one or more non-transitory, computer-readable storage media of claim 16, wherein the plurality of related failures for the network element comprises a plurality of failures that occur for the network element within a specified period of time, and wherein the specified period of time comprises a duration of a particular failure event.

18. The one or more non-transitory, computer-readable storage media of claim 16, comprising code configured to determine an impact of a failure based on network topology data representing how a plurality of network elements are communicatively connected.

19. The one or more non-transitory, computer-readable storage media of claim 16, comprising code configured to determine an impact of a failure on a link by computing a ratio of a statistical measure of an amount of traffic on the link during the failure to a statistical measure of an amount of traffic on the link before the failure.

20. The one or more non-transitory, computer-readable storage media of claim 16, wherein determining the effectiveness of network redundancies comprises computing a ratio of a statistical measure of an amount of traffic on links and devices within a network redundancy group during the failure to a statistical measure of an amount of traffic on the links and the devices within the network redundancy group before the failure.