Parallel Tracing for Performance and Detail

• US 20140019985A1
• Filed: 01/25/2013
• Published: 01/16/2014
• Est. Priority Date: 01/25/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Concurix Corporation
• Inventors: Krajec, Russell S.
• Granted Patent: US 9,207,969 B2
• Time in Patent Office: Days
• US Class Current: 718/102
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/3636 by tracing the execution of the program
• CPC Class Codes: G06F 21/44 Program or device authentication
• CPC Class Codes: G06F 21/53 by executing in a restricted environment, e.g. sandbox or secure virtual machine
• CPC Class Codes: G06F 2201/865 Monitoring of software
• CPC Class Codes: G06F 9/4806 Task transfer initiation or dispatching

Abstract

A parallel tracer may perform detailed or heavily instrumented analysis of an application in parallel with a performance or lightly instrumented version of the application. Both versions of the application may operate on the same input stream, but with the heavily instrumented version having different performance results than the lightly instrumented version. The tracing results may be used for various analyses, including optimization and debugging.

Claims

1. A method performed by a computer processor, said method comprising:

executing an application in a first environment and a second environment;

said first environment being a performance environment;

said second environment being an instrumented environment;

receiving an input stream for said application;

identifying a first unit of work within said input stream and causing said first unit of work to be executed within said first environment;

identifying a second unit of work within said input stream and causing said second unit of work to be executed within said second environment.

2. The method of claim 1 further comprising:

causing said second unit of work to be executed by said first environment.

3. The method of claim 2 further comprising:

receiving detailed trace data from said second environment for said second unit of work.

4. The method of claim 3 further comprising:

receiving performance results from said first environment for said second unit of work.

5. The method of claim 4 further comprising:

combining said performance results and said detailed trace data into a results set for said second unit of work.

6. The method of claim 5, said first environment being an instrumented environment configured with a set of performance measurements.

7. The method of claim 5, said first environment being on a first device, said second environment being on a second device.

8. The method of claim 7, said first device having hardware differences from said second device.

9. The method of claim 8, said second device being remotely located from said first device.

10. The method of claim 5, said first environment and said second environment being on a first device.

11. The method of claim 10, said first environment being operated on a first set of processors on said first device, said second environment being operated on a second set of processors on said first device.

12. The method of claim 5, said combining comprising identifying an identifier for said second unit of work in said detailed trace data and said identifier for said second unit of work in said performance trace data.

13. The method of claim 1, said performance environment comprising performance monitoring instrumentation.

14. The method of claim 1, said instrumented environment comprising process tracing instrumentation.

15. The method of claim 14, said instrumented environment further comprising memory management instrumentation.

16. The method of claim 15, said instrumented environment further comprising input/output instrumentation.

17. The method of claim 1, said first execution environment being a virtual machine.

18. The method of claim 17, said virtual machine being a process virtual machine.

19. A system comprising:

a first execution environment having a detailed tracer and executing on a first processor;

a second execution environment having a performance tracer;

a distributor that;

receives an input stream;

identifies a first unit of work within said input stream and causes said first unit of work to be executed within said first execution environment and generates a first detailed trace result;

identifies a second unit of work within said input stream and causes said second unit of work to be executed within said second execution environment and generates a first performance trace result.

20. The system of claim 19 further comprising:

a combiner that combines said first detailed trace result and said first performance trace result.