STATISTICAL PROFILING OF CLUSTER TASKS

• US 20140136698A1
• Filed: 12/19/2013
• Published: 05/15/2014
• Est. Priority Date: 12/17/2010
• Status: Abandoned Application

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Venneti, Satya, Kownacki, Ronald William, Katepalli, Sundeep, DeGraaf, John
• Application Number:
  US14/134,025
• Publication Number:
  US 20140136698A1
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: G06F 11/3003 Monitoring arrangements specially adapted to the computing system or computing system component being monitored
• CPC Class Codes: G06F 11/3082 the data filtering being achieved by aggregating or compressing the monitored data
• CPC Class Codes: G06F 11/3089 Monitoring arrangements determined by the means or processing involved in sensing the monitored data, e.g. interfaces, connectors, sensors, probes, agents software debugging using additional hardware using a specific debug interface G06F11/3656; performance evaluation by tracing or monitoring G06F11/3466
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/865 Monitoring of software
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks

Abstract

A variety of complex multiphase tasks may execute within a cluster network environment. It may be advantageous to track and analyze the processing of such multiphase tasks. Accordingly, one or more techniques and/or systems are disclosed for tracking the processing of a multiphase task within a cluster network environment. In particular, runtime support for recording task processing annotations may be embedded into a multiphase task using an analysis model. The analysis model may specify task processing annotations that are to be recorded during one or more phases of processing the multiphase task. During execution, task processing annotations may be recorded and associated with corresponding phases of the multiphase task. Task processing annotations may be aggregated into aggregated statistical task data. The aggregated statistical task data may be used to generate views of statistical data, which may provide insight into the cluster network environment (e.g., problematic nodes, bottlenecks, etc.).

Claims

1. A method for monitoring a multiphase task, comprising:

incorporating an analysis model into a multiphase task to be processed within a clustered network environment, the analysis model specifying a first task processing annotation to be recorded during a first phase of processing the multiphase task and a second task processing annotation to be recorded during a second phase of processing the multiphase task;

during the first phase of processing the multiphase task, recording the first task processing annotation; and

during the second phase of processing the multiphase task, recording the second task processing annotation.

2. The method of claim 1, the recording the first task processing annotation comprising recording the first task processing annotation on a first node within the clustered network environment, and the recording the second task processing annotation comprising recording the second task processing annotation on a second node within the clustered network environment.

3. The method of claim 1, the first phase comprising at least one of an acquire phase, a hold phase, or a commit phase, and the second phase comprising a phase different than the first phase.

4. The method of claim 1, comprising:

aggregating the first task processing annotation and the second task processing annotation with previously recorded task processing annotations to generate aggregated statistical task data.

5. The method of claim 4, comprising:

receiving a request to generate a view of statistical data corresponding to the processing of at least some of one or more multiphase tasks, the request comprising user defined statistical criteria; and

generating the view of statistical data based upon at least a portion of the aggregated statistical task data corresponding to the user defined statistical criteria.

6. The method of claim 4, the aggregating comprising:

averaging at least one of the first task processing annotation or the second task processing annotation with the previously recorded task processing annotations.

7. The method of claim 5, the view of statistical data corresponding to aggregated task processing annotations for the first phase during one or more executions of the multiphase task.

8. The method of claim 1, the recording the first task processing annotation comprising recording the first task processing annotation within a first data structure of a first remote procedure call to a first node within the clustered network environment.

9. The method of claim 8, the first data structure owned by the multiphase task.

10. The method of claim 8, the recording the second task processing annotation comprising recording the second task processing annotation within a second data structure of a second remote procedure call to a second node within the clustered network environment.

11. The method of claim 10, the first data structure and the second data structure owned by the multiphase task.

12. The method of claim 8, comprising:

utilizing a stat processing thread to process the first data structure for analysis of the first phase.

13. The method of claim 1, the first task processing annotation comprising at least one of:

a first start time of the first phase;

a first end time of the first phase;

a first processing time of the first phase on a first node within the clustered network environment;

or

first health information of the first node that processed the first phase.

14. The method of claim 13, the second task processing annotation comprising at least one of:

a second start time of the second phase;

a second end time of the second phase;

a second processing time of the second phase on a second node within the clustered network environment;

or

second health information of the second node that processed the second phase.

15. A system for monitoring a multiphase task, comprising:

an injection component configured to;

incorporate an analysis model into a multiphase task to be processed within a clustered network environment, the analysis model specifying a first task processing annotation to be recorded during a first phase of processing the multiphase task and a second task processing annotation to be recorded during a second phase of processing the multiphase task;

a recording component configured to;

during the first phase of processing the multiphase task, record the first task processing annotation; and

during the second phase of processing the multiphase task, record the second task processing annotation.

16. The system of claim 15, the recording component configured to:

record the first task processing annotation on a first node within the clustered network environment; and

record the second task processing annotation on a second node within the clustered network environment.

17. The system of claim 15, the first phase comprising at least one of an acquire phase, a hold phase, or a commit phase, and the second phase comprising a phase different than the first phase.

18. The system of claim 15, comprising:

an analysis component configured to;

aggregate the first task processing annotation and the second task processing annotation with previously recorded task processing annotations to generate aggregated statistical task data.

19. The system of claim 18, the analysis component configured to:

receive a request to generate a view of statistical data corresponding to the processing of at least some of one or more multiphase tasks, the request comprising user defined statistical criteria; and

generate the view of statistical data based upon at least a portion of the aggregated statistical task data corresponding to the user defined statistical criteria.

20. A computer readable medium comprising computer executable instructions that when executed perform a method for monitoring a multiphase task, comprising:

incorporating an analysis model into a multiphase task to be processed within a clustered network environment, the analysis model specifying a first task processing annotation to be recorded during a first phase of processing the multiphase task and a second task processing annotation to be recorded during a second phase of processing the multiphase task;

during the first phase of processing the multiphase task, recording the first task processing annotation; and

during the second phase of processing the multiphase task, recording the second task processing annotation.