Automated method and system for collecting and reporting API performance profiles

• US 20070294673A1
• Filed: 06/16/2006
• Published: 12/20/2007
• Est. Priority Date: 06/16/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Guerrera, Nicholas, Mariani, Rico, Fanning, Michael C., Gray, Jan Stephen
• Granted Patent: US 7,895,579 B2
• Time in Patent Office: Days
• US Class Current: 717/130
• 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 11/3652 in-circuit-emulation [ICE] arrangements
• CPC Class Codes: G06F 2201/865 Monitoring of software

Abstract

A method to measure the performance of an Application Programming Interface (API) includes selecting a target API and tracking memory allocation for the functional calls of the selected API to generate a list of functions called by the API. The individual functions are then called within the API in isolation and timing and memory allocation data for each are collected and recorded. The recorded results are then bucketized according to a log base 2 method and made accessible by API users to indicate the performance of specific APIs.

Claims

1. A method to measure performance of an Application Programming Interfaces (APIs), the method comprising:

(a) creating a list of APIs to measure;

(b) selecting a single target Application Programming Interface (API) from the list;

(c) intercepting calls to the target API and collecting timing and memory allocation data;

(d) recording the timing and memory allocation data for the target API, the recorded data placed into storage;

(e) merging measured results into a results table having bucketized data indicating log base 2 of the collected timing and memory allocation data; and

(f) repeating steps (b) through (e) for each API on the list.

2. The method of claim 1, further comprising:

generating a database of results and making the database available to users of API performance data.

3. The method of claim 1, further comprising:

presenting the results table via a graphical user interface using histograms for each function.

4. The method of claim 1, further comprising:

generating a list of functions called by the target API, the list comprising non-interfering functions comprising functions that never appear on the call stack at the same time;

concurrently measuring timing and memory allocation for a plurality of non-interfering functions.

5. The method of claim 4, further comprising:

merging measured results of execution of the non-interfering functions into the results table having bucketized data indicating a log base 2 of the timing and memory allocation data.

6. The method of claim 5, further comprising:

presenting the results table via a graphical user interface using histograms for each function.

7. The method of claim 1, wherein a program to measure performance of an API resides on a plurality of user computers, the method further comprising:

testing to see if the recorded data in storage exceeds a threshold amount;

selecting another API in a user computer on which to measure memory allocation and timing data if the threshold amount is not exceeded, and repeating steps (b) through (e);

submitting the recorded data in storage to a central service location if the recorded data in storage exceeds the threshold amount, the central service location having API measurement data in storage;

aggregating the submitted data with the API measurement data in storage at the central storage location; and

generating a database of results and making the database available to users of API performance data.

8. The method of claim 1, wherein selecting an API on which to measure memory allocation and timing data comprises using a random number generator to select a target API.

9. A computer system for acquiring performance data for Application Programming Interfaces (APIs), the system comprising:

a processor executing software instructions, the software instructions comprising;

a memory allocation profiler module which measures and records a number of times a function is called by a target Application Programming Interface (API) producing memory allocation measurements of the target API;

a timing profiler module which measures and records timing information related to the target API producing timing measurements of the target API;

a coordination module that runs the memory allocation module and the timing profiler module once for each invocation of the target API, wherein the coordination module avoids simultaneously measuring any two APIs which appear on the same call stack;

a merging tool module which combines the memory allocation measurements and the timing measurements of the target API into a results table; and

a user interface module which can display data of the results table in a form according to a log base 2 bucketing of the memory allocation and timing measurements.

10. The computer system of claim 9, wherein the memory allocation profiler separately measures and records inclusive and exclusive memory allocation values.

11. The computer system of claim 9, wherein the user interface module further comprises a graphical user interface that renders data of the results table into histogram form for each function call measured.

12. A computer-readable medium having computer-executable instructions for performing a method of acquiring performance data for Application Programming Interfaces (APIs), the method comprising:

(a) creating a list of APIs to measure;

(b) selecting a single target API from the list;

(c) intercepting calls to the target API and collecting timing and memory allocation data;

(d) recording the timing and memory allocation data for the target API, the recorded data placed into storage;

(e) merging measured results into a results table having bucketized data indicating a log base 2 of the collected timing and memory allocation data; and

(f) repeating steps (b) through (e) for each API on the list.

13. The computer-readable medium of claim 12, further comprising:

generating a database of results and making the database available to users of API performance data.

14. The computer-readable medium of claim 12, further comprising:

presenting the results table via a graphical user interface using histograms for each function.

15. The computer-readable medium of claim 12, further comprising:

generating a list of functions called by the target API, the list comprising non-interfering functions comprising functions that never appear on the call stack at the same time;

concurrently measuring timing and memory allocation for a plurality of non-interfering functions.

16. The computer-readable medium of claim 15, further comprising:

merging measured results of execution of the non-interfering functions into the results table having bucketized data indicating a log base 2 of the timing and memory allocation data.

17. The computer-readable medium of claim 16, further comprising:

presenting the results table via a graphical user interface using histograms for each function.

18. The computer-readable medium of claim 12, further comprising:

testing to see if the recorded data in storage exceeds a threshold amount;

selecting another API on which to measure memory allocation and timing data if the threshold amount is not exceeded, and repeating steps (b) through (e);

submitting the recorded data in storage to a central service location if the recorded data in storage exceeds the threshold amount, the central service location having API measurement data in storage;

aggregating the submitted data with the API measurement data in storage at the central storage location; and

generating a database of results and making the database available to users of API performance data.

19. The computer-readable medium of claim 12, wherein selecting an API on which to measure memory allocation and timing data comprises using a random number generator to select a target API.