Download - Scam2011 syer
![Page 1: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/1.jpg)
Identifying Performance Deviations in Thread Pools
Mark D. Syer, Bram Adams and Ahmed E. [email protected]
Software Analysis and Intelligence LabQueen’s University, Canada
1
![Page 2: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/2.jpg)
2
![Page 3: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/3.jpg)
3
4
How to detectperformancedeviations?
Needle in a haystack!
Complexarchitectures
Ultra-Large-Scale Software Systems
![Page 4: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/4.jpg)
Potential solutions must account for…
• Huge amounts of data• Limited system knowledge
4
Potential solutions should…
• Limit manual review/analysis
![Page 5: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/5.jpg)
5
4
How to detectperformancedeviations?
Hundreds of machinesThousands of threads
Needle in a haystack!
Complexarchitectures
Simulate user actions
Monitor thread behaviour by collecting resource usage metrics
![Page 6: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/6.jpg)
Simulate user actions
Monitor thread behaviour by collecting resource usage metrics
6
43
21
Group threads depending on their machine ID
![Page 7: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/7.jpg)
7
3
21
4
1
2
4
3
Group similar behaviour
Similar
Dissimilar
Find dissimilar behaviour
![Page 8: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/8.jpg)
8
AbstractionClustering
Ranking
Group threads
Find dissimilarbehaviour
Group similarbehaviour
How to group threads?
1 2
3
![Page 9: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/9.jpg)
Abstraction
Ranking
Clustering
9
Abstraction Clustering
Ranking
Machines
Threads
![Page 10: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/10.jpg)
10
does it work?
![Page 11: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/11.jpg)
11
6:00 8:004:00
Abstracting bySpace
Vs.Time
RQ1: What time period has the most deviations?
RQ2: What threads have deviations?
![Page 12: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/12.jpg)
Deviations Injected into Thread Data
12
Memory Leak Injected
![Page 13: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/13.jpg)
Deviations Injected into Thread Data
13
CPU Spike Injected
![Page 14: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/14.jpg)
Most Deviations are Identified
Precision Recall
Top Level 100% 100%
Thread Level 100% 76.61%
14
1-recall 2-recall 3-recall
86.67% 96.67% 100%
![Page 15: Scam2011 syer](https://reader036.vdocuments.pub/reader036/viewer/2022062412/58edc5931a28ab303a8b477f/html5/thumbnails/15.jpg)
15