Abstract

This paper describes a measurement-based performability model based on error and resource usage data collected on a multiprocessor system. A method for identifying the model structure is introduced and the resulting model is validated against real data. Model development from the collection of raw data to the estimation of the expected reward is described. Both normal and error behavior of the system are characterized. The measured data show that the holding times in key operational and error states are not simple exponentials and that a semi-Markov process is necessary to model the system behavior. A reward function, based on the service rate and the error rate in each state, is then defined in order to estimate the performability of the system and to depict the cost of different types of errors.

Original languageEnglish (US)
Pages (from-to)478-484
Number of pages7
JournalIEEE Transactions on Computers
Volume37
Issue number4
DOIs
StatePublished - Apr 1988

Fingerprint

Performability
Reward
Modeling
Semi-Markov Process
Multiprocessor Systems
Model
Error Rate
Model structures
Markov processes
Model-based
Resources
Necessary
Costs
Estimate

Keywords

  • Error
  • measurements
  • performability
  • semi-Markov
  • workload

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computational Theory and Mathematics

Cite this

Performability Modeling Based on Real Data : A Case Study. / Hsueh, M. C.; Iyer, R. K.

In: IEEE Transactions on Computers, Vol. 37, No. 4, 04.1988, p. 478-484.

Research output: Contribution to journalArticle

@article{a306e194b06e4a1c8d545c00c8b17903,
title = "Performability Modeling Based on Real Data: A Case Study",
abstract = "This paper describes a measurement-based performability model based on error and resource usage data collected on a multiprocessor system. A method for identifying the model structure is introduced and the resulting model is validated against real data. Model development from the collection of raw data to the estimation of the expected reward is described. Both normal and error behavior of the system are characterized. The measured data show that the holding times in key operational and error states are not simple exponentials and that a semi-Markov process is necessary to model the system behavior. A reward function, based on the service rate and the error rate in each state, is then defined in order to estimate the performability of the system and to depict the cost of different types of errors.",
keywords = "Error, measurements, performability, semi-Markov, workload",
author = "Hsueh, {M. C.} and Iyer, {R. K.}",
year = "1988",
month = "4",
doi = "10.1109/12.2195",
language = "English (US)",
volume = "37",
pages = "478--484",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "4",

}

TY - JOUR

T1 - Performability Modeling Based on Real Data

T2 - A Case Study

AU - Hsueh, M. C.

AU - Iyer, R. K.

PY - 1988/4

Y1 - 1988/4

N2 - This paper describes a measurement-based performability model based on error and resource usage data collected on a multiprocessor system. A method for identifying the model structure is introduced and the resulting model is validated against real data. Model development from the collection of raw data to the estimation of the expected reward is described. Both normal and error behavior of the system are characterized. The measured data show that the holding times in key operational and error states are not simple exponentials and that a semi-Markov process is necessary to model the system behavior. A reward function, based on the service rate and the error rate in each state, is then defined in order to estimate the performability of the system and to depict the cost of different types of errors.

AB - This paper describes a measurement-based performability model based on error and resource usage data collected on a multiprocessor system. A method for identifying the model structure is introduced and the resulting model is validated against real data. Model development from the collection of raw data to the estimation of the expected reward is described. Both normal and error behavior of the system are characterized. The measured data show that the holding times in key operational and error states are not simple exponentials and that a semi-Markov process is necessary to model the system behavior. A reward function, based on the service rate and the error rate in each state, is then defined in order to estimate the performability of the system and to depict the cost of different types of errors.

KW - Error

KW - measurements

KW - performability

KW - semi-Markov

KW - workload

UR - http://www.scopus.com/inward/record.url?scp=0023999829&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0023999829&partnerID=8YFLogxK

U2 - 10.1109/12.2195

DO - 10.1109/12.2195

M3 - Article

AN - SCOPUS:0023999829

VL - 37

SP - 478

EP - 484

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 4

ER -