DISTRIBUTED AGGREGATION OF REAL-TIME METRICS FOR LARGE SCALE DISTRIBUTED SYSTEMS

• US 20140032741A1
• Filed: 07/27/2012
• Published: 01/30/2014
• Est. Priority Date: 07/27/2012
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Lau, Cheuk Wan William, Joy, George, Cherukuri, Ravikant, Yamini, Smriti
• Granted Patent: US 10,075,520 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 2201/835 Timestamp
• CPC Class Codes: G06F 2201/865 Monitoring of software
• CPC Class Codes: H04L 43/065 related to network devices
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 67/125 involving control of end-device applications over a network
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34
• CPC Class Codes: H04L 69/28 Timers or timing mechanisms used in protocols

Abstract

In one embodiment, a metric collection system may collect application metrics across multiple data centers in specific time intervals. A local collection point 108 may establish an absolute reference time for a local data center 102. The local collection point 108 may determine a source reporting interval for an application 106 based on the absolute reference time. The local collection point 108 may receive a source metric report 300 for the application 106 based on the source reporting interval.

Claims

1. A machine-implemented method, comprising:

establishing an absolute reference time for a local data center at a local collection point;

determining a source reporting interval for an application based on the absolute reference time; and

receiving a source metric report for the application based on the source reporting interval.

2. The method of claim 1, further comprising:

assigning the source reporting interval to the source metric report when the source metric report is received during a reporting window.

3. The method of claim 1, further comprising:

processing the source metric report.

4. The method of claim 1, further comprising:

selecting a source action based on the source metric report.

5. The method of claim 1, further comprising:

compiling a local batch metric report to send to a remote data center.

6. The method of claim 1, further comprising:

determining a batch interval for sending a local batch metric report to a remote data center.

7. The method of claim 1, further comprising:

calculate a batch offset for the batch interval by applying a batch offset hash to an application identifier.

8. The method of claim 1, further comprising:

identifying a batch publishing trigger for sending a local batch metric report.

9. The method of claim 1, further comprising:

sending a local batch metric report to a remote collection point for a remote data center.

10. The method of claim 1, further comprising:

determine a remote collection point by applying a collection hash to an application identifier.

11. The method of claim 1, further comprising:

sending a local batch metric report to a remote proxy to determine a remote collection point for a remote data center.

12. The method of claim 1, further comprising:

receiving a remote batch metric report chronicling the application at a remote data center.

13. The method of claim 1, further comprising:

selecting a remote action based on a remote batch metric report.

14. The method of claim 1, further comprising:

sending a time synchronization signal to a remote data center.

15. A tangible machine-readable medium having a set of instructions detailing a method stored thereon that when executed by one or more processors cause the one or more processors to perform the method, the method comprising:

establishing an absolute reference time for a local data center at a source local data center server;

determining a source reporting interval for an application based on the absolute reference time; and

sending a source metric report for the application for the source reporting interval.

16. The tangible machine-readable medium of claim 15, wherein the method further comprises:

calculating a source offset for a source reporting interval by applying a source offset hash to an application identifier.

17. The tangible machine-readable medium of claim 15, wherein the method further comprises:

determining the local collection point by applying a collection hash to an application identifier.

18. The tangible machine-readable medium of claim 15, wherein the method further comprises:

identifying a source publishing trigger to send the source metric report.

19. A data center server, comprising:

a communication interface that receives a source metric report for an application during a reporting window; and

a processor that assigns a source reporting interval based on an absolute reference time to the source metric report.

20. The data center server of claim 19, wherein the processor compiles a local batch metric report to send to a remote data center at a batch interval.