Benchmarking: You're Doing It Wrong (StrangeLoop 2014)
•
7 j'aime•2,531 vues
Knowledge of how to set up good benchmarks is invaluable in understanding performance of the system. Writing correct and useful benchmarks is hard, and verification of the results is difficult and prone to errors. When done right, benchmarks guide teams to improve the performance of their systems. When done wrong, hours of effort may result in a worse performing application, upset customers or worse! In this talk, we will discuss what you need to know to write better benchmarks. We will look at examples of bad benchmarks and learn about what biases can invalidate the measurements, in the hope of correctly applying our new-found skills and avoiding such pitfalls in the future.
Recommandé
Recommandé
Contenu connexe
Tendances
Tendances (20)
Similaire à Benchmarking: You're Doing It Wrong (StrangeLoop 2014)
Similaire à Benchmarking: You're Doing It Wrong (StrangeLoop 2014) (20)
Plus de Aysylu Greenberg
Plus de Aysylu Greenberg (20)
Dernier
Dernier (20)
Benchmarking: You're Doing It Wrong (StrangeLoop 2014)
- 1. Benchmarking: You’re Doing It Wrong Aysylu Greenberg @aysylu22
- 2. In Memes…
- 3. To Write Good Benchmarks… Need to be Full Stack
- 4. What’s a Benchmark How fast? Your process vs Goal Your process vs Best PracLces
- 5. Today • How Not to Write Benchmarks • Benchmark Setup & Results: - Wrong about the machine - Wrong about stats - Wrong about what maOers • Becoming Less Wrong
- 6. HOW NOT TO WRITE BENCHMARKS
- 7. Website Serving Images • Access 1 image 1000 Lmes • Latency measured for each access • Start measuring immediately • 3 runs • Find mean • Dev machine Web Request Server Cache S3
- 8. WHAT’S WRONG WITH THIS BENCHMARK?
- 9. You’re wrong about the machine BENCHMARK SETUP & RESULTS: COMMON PITFALLS
- 10. Wrong About the Machine • Cache, cache, cache, cache!
- 11. It’s Caches All The Way Down
- 12. It’s Caches All The Way Down
- 13. Caches in Benchmarks Prof. Saman Amarasinghe, MIT 2009
- 14. Caches in Benchmarks Prof. Saman Amarasinghe, MIT 2009
- 15. Caches in Benchmarks Prof. Saman Amarasinghe, MIT 2009
- 16. Caches in Benchmarks Prof. Saman Amarasinghe, MIT 2009
- 17. Caches in Benchmarks Prof. Saman Amarasinghe, MIT 2009
- 18. Website Serving Images • Access 1 image 1000 Lmes • Latency measured for each access • Start measuring immediately • 3 runs • Find mean • Dev machine Web Request Server Cache S3
- 19. Wrong About the Machine • Cache, cache, cache, cache! • Warmup & Timing
- 20. Website Serving Images • Access 1 image 1000 Lmes • Latency measured for each access • Start measuring immediately • 3 runs • Find mean • Dev machine Web Request Server Cache S3
- 21. Wrong About the Machine • Cache, cache, cache, cache! • Warmup & Timing • Periodic interference
- 22. Website Serving Images • Access 1 image 1000 Lmes • Latency measured for each access • Start measuring immediately • 3 runs • Find mean • Dev machine Web Request Server Cache S3
- 23. Wrong About the Machine • Cache, cache, cache, cache! • Warmup & Timing • Periodic interference • Different specs in test vs prod machines
- 24. Website Serving Images • Access 1 image 1000 Lmes • Latency measured for each access • Start measuring immediately • 3 runs • Find mean • Dev machine Web Request Server Cache S3
- 25. Wrong About the Machine • Cache, cache, cache, cache! • Warmup & Timing • Periodic interference • Different specs in test vs prod machines • Power mode changes
- 26. You’re wrong about the stats BENCHMARK SETUP & RESULTS: COMMON PITFALLS
- 27. Wrong About Stats • Too few samples
- 28. Wrong About Stats 120 100 80 60 40 20 0 Convergence of Median on Samples 0 10 20 30 40 50 60 Latency Time Stable Samples Stable Median Decaying Samples Decaying Median
- 29. Website Serving Images • Access 1 image 1000 Lmes • Latency measured for each access • Start measuring immediately • 3 runs • Find mean • Dev machine Web Request Server Cache S3
- 30. Wrong About Stats • Too few samples • Non-‐Gaussian
- 31. Website Serving Images • Access 1 image 1000 Lmes • Latency measured for each access • Start measuring immediately • 3 runs • Find mean • Dev machine Web Request Server Cache S3
- 32. Wrong About Stats • Too few samples • Non-‐Gaussian • MulLmodal distribuLon
- 33. MulLmodal DistribuLon 50% 99% # occurrences Latency 5 ms 10 ms
- 34. Wrong About Stats • Too few samples • Non-‐Gaussian • MulLmodal distribuLon • Outliers
- 35. You’re wrong about what maOers BENCHMARK SETUP & RESULTS: COMMON PITFALLS
- 36. Wrong About What MaOers • Premature opLmizaLon
- 37. “Programmers waste enormous amounts of Lme thinking about … the speed of noncriLcal parts of their programs ... Forget about small efficiencies …97% of the Lme: premature opKmizaKon is the root of all evil. Yet we should not pass up our opportuniLes in that criLcal 3%.” -‐-‐ Donald Knuth
- 38. Wrong About What MaOers • Premature opLmizaLon • UnrepresentaLve Workloads
- 39. Wrong About What MaOers • Premature opLmizaLon • UnrepresentaLve Workloads • Memory pressure
- 40. The How BECOMING LESS WRONG
- 41. Becoming Less Wrong User AcLons MaOer X > Y for workload Z with trade offs A, B, and C -‐ hOp://www.toomuchcode.org/
- 42. Becoming Less Wrong Profiling Code InstrumentaLon Aggregate Over Logs Traces
- 43. Microbenchmarking: Blessing & Curse + Quick & cheap + Answers narrow ?s well - Open misleading results - Not representaLve of the program
- 44. Microbenchmarking: Blessing & Curse • Choose your N wisely
- 45. Choose Your N Wisely Prof. Saman Amarasinghe, MIT 2009
- 46. Microbenchmarking: Blessing & Curse • Choose your N wisely • Measure side effects
- 47. Microbenchmarking: Blessing & Curse • Choose your N wisely • Measure side effects • Beware of clock resoluLon
- 48. Microbenchmarking: Blessing & Curse • Choose your N wisely • Measure side effects • Beware of clock resoluLon • Dead Code EliminaLon
- 49. Microbenchmarking: Blessing & Curse • Choose your N wisely • Measure side effects • Beware of clock resoluLon • Dead Code EliminaLon • Constant work per iteraLon
- 50. Non-‐Constant Work Per IteraLon
- 51. Follow-‐up Material • How NOT to Measure Latency by Gil Tene – hOp://www.infoq.com/presentaLons/latency-‐piralls • Taming the Long Latency Tail on highscalability.com – hOp://highscalability.com/blog/2012/3/12/google-‐taming-‐the-‐ long-‐latency-‐tail-‐when-‐more-‐machines-‐equal.html • Performance Analysis Methodology by Brendan Gregg – hOp://www.brendangregg.com/methodology.html • Robust Java benchmarking by Brent Boyer – hOp://www.ibm.com/developerworks/library/j-‐benchmark1/ – hOp://www.ibm.com/developerworks/library/j-‐benchmark2/ • Benchmarking arLcles by Aleksey Shipilëv – hOp://shipilev.net/#benchmarking
- 55. Benchmarking: You’re Doing It Wrong Aysylu Greenberg @aysylu22