Synthesizing representative I/O workloads for TPC-H

Jianyong Zhang; Anand Sivasubramaniam; Hubertus Franke; Natarajan Gautam; Yanyong Zhang; Shailabh Nagar

Synthesizing representative I/O workloads for TPC-H

Jianyong Zhang, Anand Sivasubramaniam, Hubertus Franke, Natarajan Gautam, Yanyong Zhang, Shailabh Nagar

Research output: Contribution to journal › Conference Article › peer-review

Abstract

Synthesizing I/O requests that can accurately capture workload behavior is extremely valuable for the design, implementation and optimization of disk subsystems. This paper presents a synthetic workload generator for TPC-H, an important decision-support commercial workload, by completely characterizing the arrival and access patterns of its queries. We present a novel approach for parameterizing the behavior of inter-mingling streams of sequential requests, and exploit correlations between multiple attributes of these requests, to generate disk block-level traces that are shown to accurately mimic the behavior of a real trace in terms of response time characteristics for each TPC-H query.

Original language	English (US)
Pages (from-to)	142-151
Number of pages	10
Journal	IEEE High-Performance Computer Architecture Symposium Proceedings
Volume	10
State	Published - 2004
Externally published	Yes
Event	Proceedings - 10th International Symposium on High Performance Computer Architecture - Madrid, Spain Duration: Feb 14 2004 → Feb 18 2004

ASJC Scopus subject areas

Hardware and Architecture

Cite this

@article{64a4e3ecfce44beb8f8c4b3cd2324dad,

title = "Synthesizing representative I/O workloads for TPC-H",

abstract = "Synthesizing I/O requests that can accurately capture workload behavior is extremely valuable for the design, implementation and optimization of disk subsystems. This paper presents a synthetic workload generator for TPC-H, an important decision-support commercial workload, by completely characterizing the arrival and access patterns of its queries. We present a novel approach for parameterizing the behavior of inter-mingling streams of sequential requests, and exploit correlations between multiple attributes of these requests, to generate disk block-level traces that are shown to accurately mimic the behavior of a real trace in terms of response time characteristics for each TPC-H query.",

author = "Jianyong Zhang and Anand Sivasubramaniam and Hubertus Franke and Natarajan Gautam and Yanyong Zhang and Shailabh Nagar",

year = "2004",

language = "English (US)",

volume = "10",

pages = "142--151",

journal = "IEEE High-Performance Computer Architecture Symposium Proceedings",

issn = "1530-0897",

publisher = "IEEE Computer Society",

note = "Proceedings - 10th International Symposium on High Performance Computer Architecture ; Conference date: 14-02-2004 Through 18-02-2004",

}

TY - JOUR

T1 - Synthesizing representative I/O workloads for TPC-H

AU - Zhang, Jianyong

AU - Sivasubramaniam, Anand

AU - Franke, Hubertus

AU - Gautam, Natarajan

AU - Zhang, Yanyong

AU - Nagar, Shailabh

PY - 2004

Y1 - 2004

N2 - Synthesizing I/O requests that can accurately capture workload behavior is extremely valuable for the design, implementation and optimization of disk subsystems. This paper presents a synthetic workload generator for TPC-H, an important decision-support commercial workload, by completely characterizing the arrival and access patterns of its queries. We present a novel approach for parameterizing the behavior of inter-mingling streams of sequential requests, and exploit correlations between multiple attributes of these requests, to generate disk block-level traces that are shown to accurately mimic the behavior of a real trace in terms of response time characteristics for each TPC-H query.

AB - Synthesizing I/O requests that can accurately capture workload behavior is extremely valuable for the design, implementation and optimization of disk subsystems. This paper presents a synthetic workload generator for TPC-H, an important decision-support commercial workload, by completely characterizing the arrival and access patterns of its queries. We present a novel approach for parameterizing the behavior of inter-mingling streams of sequential requests, and exploit correlations between multiple attributes of these requests, to generate disk block-level traces that are shown to accurately mimic the behavior of a real trace in terms of response time characteristics for each TPC-H query.

UR - http://www.scopus.com/inward/record.url?scp=2342525809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2342525809&partnerID=8YFLogxK

M3 - Conference Article

AN - SCOPUS:2342525809

SN - 1530-0897

VL - 10

SP - 142

EP - 151

JO - IEEE High-Performance Computer Architecture Symposium Proceedings

JF - IEEE High-Performance Computer Architecture Symposium Proceedings

T2 - Proceedings - 10th International Symposium on High Performance Computer Architecture

Y2 - 14 February 2004 through 18 February 2004

ER -

Synthesizing representative I/O workloads for TPC-H

Abstract

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this