Diagnosing root causes of system level performance violations

Lingyi Liu, Xuanyu Zhong, Xiaotao Chen, Shobha Vasudevan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Diagnosing performance violations is one of the biggest challenges in transaction level modeling of systems. In this paper, we propose a methodology to localize root causes of latency or throughput violations. We present a concurrent pattern mining approach to infer frequent patterns from transaction traces to localize root causes. We apply three categories of domain knowledge from the violation and models to filter the irrelevant transaction traces and increase the effectiveness of the mining results. We provide three culprit scenarios to mining algorithm by including transaction traces relevant to the corresponding culprit scenario. The mined concurrent patterns then belong to that culprit scenario. We provide a case study for diagnosing performance violations of an experimental platform and show that our domain knowledge can reduce the number of transaction traces by up to 92.8%. The concurrent pattern mining pinpoints the root cause to one of fewer than 10 patterns among 100000 transaction traces.

Original languageEnglish (US)
Title of host publication2013 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - Digest of Technical Papers
Pages295-302
Number of pages8
DOIs
StatePublished - Dec 1 2013
Event2013 32nd IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - San Jose, CA, United States
Duration: Nov 18 2013Nov 21 2013

Publication series

NameIEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD
ISSN (Print)1092-3152

Other

Other2013 32nd IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013
CountryUnited States
CitySan Jose, CA
Period11/18/1311/21/13

Fingerprint

Throughput

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design

Cite this

Liu, L., Zhong, X., Chen, X., & Vasudevan, S. (2013). Diagnosing root causes of system level performance violations. In 2013 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - Digest of Technical Papers (pp. 295-302). [6691135] (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD). https://doi.org/10.1109/ICCAD.2013.6691135

Diagnosing root causes of system level performance violations. / Liu, Lingyi; Zhong, Xuanyu; Chen, Xiaotao; Vasudevan, Shobha.

2013 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - Digest of Technical Papers. 2013. p. 295-302 6691135 (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Liu, L, Zhong, X, Chen, X & Vasudevan, S 2013, Diagnosing root causes of system level performance violations. in 2013 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - Digest of Technical Papers., 6691135, IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD, pp. 295-302, 2013 32nd IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013, San Jose, CA, United States, 11/18/13. https://doi.org/10.1109/ICCAD.2013.6691135
Liu L, Zhong X, Chen X, Vasudevan S. Diagnosing root causes of system level performance violations. In 2013 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - Digest of Technical Papers. 2013. p. 295-302. 6691135. (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD). https://doi.org/10.1109/ICCAD.2013.6691135
Liu, Lingyi ; Zhong, Xuanyu ; Chen, Xiaotao ; Vasudevan, Shobha. / Diagnosing root causes of system level performance violations. 2013 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - Digest of Technical Papers. 2013. pp. 295-302 (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD).
@inproceedings{9a9806f7c29f4cfea6c5867668ae5f40,
title = "Diagnosing root causes of system level performance violations",
abstract = "Diagnosing performance violations is one of the biggest challenges in transaction level modeling of systems. In this paper, we propose a methodology to localize root causes of latency or throughput violations. We present a concurrent pattern mining approach to infer frequent patterns from transaction traces to localize root causes. We apply three categories of domain knowledge from the violation and models to filter the irrelevant transaction traces and increase the effectiveness of the mining results. We provide three culprit scenarios to mining algorithm by including transaction traces relevant to the corresponding culprit scenario. The mined concurrent patterns then belong to that culprit scenario. We provide a case study for diagnosing performance violations of an experimental platform and show that our domain knowledge can reduce the number of transaction traces by up to 92.8{\%}. The concurrent pattern mining pinpoints the root cause to one of fewer than 10 patterns among 100000 transaction traces.",
author = "Lingyi Liu and Xuanyu Zhong and Xiaotao Chen and Shobha Vasudevan",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/ICCAD.2013.6691135",
language = "English (US)",
isbn = "9781479910717",
series = "IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD",
pages = "295--302",
booktitle = "2013 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - Digest of Technical Papers",

}

TY - GEN

T1 - Diagnosing root causes of system level performance violations

AU - Liu, Lingyi

AU - Zhong, Xuanyu

AU - Chen, Xiaotao

AU - Vasudevan, Shobha

PY - 2013/12/1

Y1 - 2013/12/1

N2 - Diagnosing performance violations is one of the biggest challenges in transaction level modeling of systems. In this paper, we propose a methodology to localize root causes of latency or throughput violations. We present a concurrent pattern mining approach to infer frequent patterns from transaction traces to localize root causes. We apply three categories of domain knowledge from the violation and models to filter the irrelevant transaction traces and increase the effectiveness of the mining results. We provide three culprit scenarios to mining algorithm by including transaction traces relevant to the corresponding culprit scenario. The mined concurrent patterns then belong to that culprit scenario. We provide a case study for diagnosing performance violations of an experimental platform and show that our domain knowledge can reduce the number of transaction traces by up to 92.8%. The concurrent pattern mining pinpoints the root cause to one of fewer than 10 patterns among 100000 transaction traces.

AB - Diagnosing performance violations is one of the biggest challenges in transaction level modeling of systems. In this paper, we propose a methodology to localize root causes of latency or throughput violations. We present a concurrent pattern mining approach to infer frequent patterns from transaction traces to localize root causes. We apply three categories of domain knowledge from the violation and models to filter the irrelevant transaction traces and increase the effectiveness of the mining results. We provide three culprit scenarios to mining algorithm by including transaction traces relevant to the corresponding culprit scenario. The mined concurrent patterns then belong to that culprit scenario. We provide a case study for diagnosing performance violations of an experimental platform and show that our domain knowledge can reduce the number of transaction traces by up to 92.8%. The concurrent pattern mining pinpoints the root cause to one of fewer than 10 patterns among 100000 transaction traces.

UR - http://www.scopus.com/inward/record.url?scp=84893365228&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893365228&partnerID=8YFLogxK

U2 - 10.1109/ICCAD.2013.6691135

DO - 10.1109/ICCAD.2013.6691135

M3 - Conference contribution

SN - 9781479910717

T3 - IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD

SP - 295

EP - 302

BT - 2013 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2013 - Digest of Technical Papers

ER -