Platform matching systems and methods

• US 8,972,223 B2
• Filed: 06/19/2012
• Issued: 03/03/2015
• Est. Priority Date: 05/09/2008
• Status: Active Grant

• Current Assignee: Credit Suisse Securities USA LLC (Credit Suisse)
• Original Assignee: Credit Suisse Securities USA LLC (Credit Suisse)
• Inventors: Yatko, Steven W., Pfeffer, Warren M., Murphy, Killian D.
• Primary Examiner(s): Desta, Elias
• Application Number:
  US13/526,703
• Publication Number:
  US 20130124145A1
• Time in Patent Office: 987 Days
• Field of Search: 702/123, 702/186, 702/189, 717100-106, 717120-124, 717/140, 717/203
• US Class Current: 702/186
• CPC Class Codes: G06F 11/3428 Benchmarking
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/5044 considering hardware capabilities

Abstract

Pursuant to some embodiments, a platform matching system, method, apparatus and means are provided which include identifying at least a first benchmark test procedure to evaluate at least a first and a second target platform for use with an application, the at least first benchmark test procedure including at least a first unit test; receiving test result data from tests performed on the at least first and second target platform using the at least first benchmark test procedure; and performing a matching analysis to determine which of the at least first and second target platform is a best fit for the application.

Claims

1. A method, comprising:

identifying, by a processor, at least a first benchmark test procedure to evaluate at least a first and a second target platform for use with an application, the at least first benchmark test procedure including at least a first unit test;

receiving, by a communication device, test result data from tests performed on the at least first and second target platform using the at least first benchmark test procedure; and

performing, by the processor, a matching analysis to determine which of the at least first and second target platform is a best fit for the application;

wherein the identifying, by the processor, at least a first benchmark test procedure further includes generating an application resource usage profile associated with the application; and

wherein the application resource usage profile includes a plurality of canonical forms, each canonical form associated with a relevance weighting.

2. The method of claim 1, wherein the plurality of canonical forms include at least one canonical form selected from the group consisting of:

multithreaded, networking, floating point, memory intensive, integer intensive, and I/O disk processing.

3. The method of claim 1, wherein the benchmark test procedure is selected from a library of available benchmark test procedures.

4. The method of claim 3, wherein the benchmark test procedure comprises a plurality of unit test procedures, and the test result data includes test result data from each of the plurality of unit test procedures.

5. A method comprising:

identifying, by a processor, at least a first benchmark test procedure to evaluate at least a first and a second target platform for use with an application, the at least first benchmark test procedure including at least a first unit test;

receiving, by a communication device, test result data from tests performed on the at least first and second target platform using the at least first benchmark test procedure; and

performing, by the processor, a matching analysis to determine which of the at least first and second target platform is a best fit for the application;

wherein the matching analysis by the processor further comprises;

grouping the test result data by target platform, by benchmark test, and by unit test;

generating a set of platform placement scores within each benchmark test; and

generating a set of platform rankings based at least on the platform placement scores.

6. The method of claim 5, wherein the matching analysis by the processor further comprises normalizing the set of platform placement scores within each benchmark test.

7. A processing apparatus, comprising:

a communication device to receive and transmit information via a communication network;

a processor coupled to the communication device; and

a storage device in communication with the processor and storing instructions adapted to be executed by the processor to;

identify at least a first benchmark test procedure to evaluate at least a first and a second target platform for use with an application, the at least first benchmark test procedure including at least a first unit test;

receive test result data from tests performed on the at least first and second target platform using the at least first benchmark test procedure; and

perform a matching analysis to determine which of the at least first and second target platform is a best fit for the application;

wherein the instructions adapted to identify at least a first benchmark test procedure further include instructions to generate an application resource usage profile associated with the application;

and wherein the application resource usage profile includes a plurality of canonical forms, each canonical form associated with a relevance weighting.

8. The method of claim 7, wherein the plurality of canonical forms include at least one canonical form selected from the group consisting of:

multithreaded, networking, floating point, memory intensive, integer intensive, and I/O disk processing.

9. The apparatus of claim 7, wherein the application resource usage profile includes at least one canonical form associated with the application.

10. The apparatus of claim 7, wherein the benchmark test procedure is selected from a library of available benchmark test procedures.

11. The apparatus of claim 10, wherein the benchmark test procedure comprises a plurality of unit test procedures, and the test result data includes test result data from each of the plurality of unit test procedures.

12. A processing apparatus comprising:

a communication device to receive and transmit information via a communication network;

a processor coupled to the communication device; and

a storage device in communication with the processor and storing instructions adapted to be executed by the processor to;

identify at least a first benchmark test procedure to evaluate at least a first and a second target platform for use with an application, the at least first benchmark test procedure including at least a first unit test;

generate an application resource usage profile associated with the application, wherein the application resource usage profile includes a plurality of canonical forms, each canonical form associated with a relevance weighting;

receive test result data from tests performed on the at least first and second target platform using the at least first benchmark test procedure; and

perform a matching analysis to determine which of the at least first and second target platform is a best fit for the application.

13. The apparatus of claim 12, wherein the instructions adapted to perform a matching analysis further comprise instructions adapted to:

group the test result data by target platform, by benchmark test, and by unit test;

generate a set of platform placement scores within each benchmark test;

normalize the set of platform placement scores within each benchmark test; and

generate a set of platform rankings.

14. A platform matching system, comprising:

an application resource profile generator to;

(i) monitor an application during runtime, (ii) generate data identifying the time spent in resource access system calls, (iii) generate an application resource profile identifying a ratio of time spent in the resource access system calls to the overall application runtime.

15. The system of claim 14, further comprising a platform analysis system to:

(i) identify a set of benchmark tests;

(ii) receive benchmark test data associated with one or more target platforms; and

(iii) store the benchmark test data.

16. The system of claim 15, further comprising a matching system to:

(i) receive the application resource profile;

(ii) receive the benchmark test data associated with at least a first target platform;

(iii) compare the application resource profile with the benchmark test data; and

(iv) match the application with at least a first target platform based on the comparison.

17. The system of claim 16, further comprising a monitoring system to:

(i) receive production runtime data from a plurality of applications deployed in a production environment; and

(ii) store the runtime data for use by the matching system.

18. A platform matching system, comprising:

an application resource profile generator configured to monitor an application during runtime and generate an application resource profile, wherein the application resource usage profile includes a plurality of canonical forms, each canonical form associated with a relevance weighting;

a platform analysis system configured to receive benchmark test data associated with one or more target platforms; and

a matching system configured to compare the application resource profile with the benchmark test data, and match the application with at least one of the one or more target platforms based on the comparison.

20. The system of claim 18, further comprising a monitoring system to receive production runtime data from a plurality of applications deployed in a production environment.

19. A platform matching system, comprising:

an application resource profile generator configured to monitor an application during runtime and generate an application resource profile by identifying a ratio of time spent in one or more resource access system calls to an overall application runtime;

a platform analysis system configured to receive benchmark test data associated with one or more target platforms; and

a matching system configured to compare the application resource profile with the benchmark test data, and match the application with at least one of the one or more target platforms based on the comparison.