Instrumenting the MG applicaiton of NAS Parallel Benchmark

573 views

Published on

Course: Measurement Tools and Techniques (10-15min): Instrumenting the MG applicaiton of NAS Parallel Benchmark

Aim of this presentation: Show the steps followed for instrumenting the application.

Published in: Technology, Business
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
573
On SlideShare
0
From Embeds
0
Number of Embeds
6
Actions
Shares
0
Downloads
2
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Instrumenting the MG applicaiton of NAS Parallel Benchmark

  1. 1. 34324 - Measurement Tools and Techniques Instrumenting the MG application of NAS Parallel Benchmark Maria Stylianou marsty5@gmail.com 20-APR-2012
  2. 2. Outline● Basic Information● Instrumentation ● By observation ● Using Performance Counters ● Using Histograms● Conclusions 2
  3. 3. Outline● Basic Information● Instrumentation ● By observation ● Using Performance Counters ● Using Histograms● Conclusions 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 3 (c) with Histograms, 3 - Conclusions
  4. 4. Basic InformationExecution Environments● Personal Laptop ● Ubuntu 11.10, 64-bit ● Intel Quad Core i5 ● 4GB RAM● Boada Server ● Intel(R) Xeon(R) CPU E5645 @ 2.40GHz ● 12 Cores with HT support ● 24 GΒ RAM 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 4 (c) with Histograms, 3 - Conclusions
  5. 5. Basic InformationNAS Parallel Benchmark● Evaluate the performance of parallel supercomputers● Several Applications MG – MPI Version ● IS, EP, CG, MG Multi-Grid on a sequence ● FT, BT, SP, LU of meshes● Extrae → Produce traces● Paraver → Analyse traces 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 5 (c) with Histograms, 3 - Conclusions
  6. 6. Outline● Basic Information● Instrumentation ● By observation ● Using Performance Counters ● Using Histograms● Conclusions 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 6 (c) with Histograms, 3 - Conclusions
  7. 7. Instrumentation by Observation1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 7 (c) with Histograms, 3 - Conclusions
  8. 8. Instrumentation by Observation1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 8 (c) with Histograms, 3 - Conclusions
  9. 9. Instrumentation by Observation1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 9 (c) with Histograms, 3 - Conclusions
  10. 10. Instrumentation by ObservationInitialization 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 10 (c) with Histograms, 3 - Conclusions
  11. 11. Instrumentation by ObservationExecution 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 11 (c) with Histograms, 3 - Conclusions
  12. 12. Instrumentation by ObservationFinalization 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 12 (c) with Histograms, 3 - Conclusions
  13. 13. Outline● Basic Information● Instrumentation ● By observation ● Using Performance Counters ● Using Histograms● Conclusions 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 13 (c) with Histograms, 3 - Conclusions
  14. 14. Instrumentation using Performance CountersInstructions 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 14 (c) with Histograms, 3 - Conclusions
  15. 15. Instrumentation using Performance CountersCycles 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 15 (c) with Histograms, 3 - Conclusions
  16. 16. Instrumentation using Performance CountersIPC: Instructions Per Cycle 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 16 (c) with Histograms, 3 - Conclusions
  17. 17. Instrumentation using Performance CountersL1 Cache Misses 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 17 (c) with Histograms, 3 - Conclusions
  18. 18. Outline● Basic Information● Instrumentation ● By observation ● Using Performance Counters ● Using Histograms● Conclusions 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 18 (c) with Histograms, 3 - Conclusions
  19. 19. Instrumentation using HistogramsTimeHistogram 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 19 (c) with Histograms, 3 - Conclusions
  20. 20. Instrumentation using HistogramsPercentageHistogram 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 20 (c) with Histograms, 3 - Conclusions
  21. 21. Instrumentation using HistogramsPercentageHistogram 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 21 (c) with Histograms, 3 - Conclusions
  22. 22. Instrumentation using HistogramsPercentageHistogram 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 22 (c) with Histograms, 3 - Conclusions
  23. 23. Conclusions● Scalability ● In laptop: No way! ● In Boada: Yes!● #Processors Increase → L1 Cache Misses Increase● Useful information very fast → Histograms! 1 - Basic Info, 2 - Instrumentation (a) by Observation, (b) with Performance Counters, 23 (c) with Histograms, 3 - Conclusions
  24. 24. 34324 - Measurement Tools and Techniques Instrumenting the MG application of NAS Parallel Benchmark Maria Stylianou marsty5@gmail.com 20-APR-2012 24

×