Critical path based microarchitectural bottleneck analysis for out-of-order execution

Research output: Contribution to journalArticle

Abstract

SUMMARY Correctly understanding microarchitectural bottlenecks is important to optimize performance and energy of OoO (Out-of-Order) processors. Although CPI (Cycles Per Instruction) stack has been utilized for this purpose, it stacks architectural events heuristically by counting how many times the events occur, and the order of stacking affects the result, which may be misleading. It is because CPI stack does not consider the execution path of dynamic instructions. Critical path analysis (CPA) is a well-known method to identify the critical execution path of dynamic instruction execution on OoO processors. The critical path consists of the sequence of events that determines the execution time of a program on a certain processor. We develop a novel representation of CPCI stack (Cycles Per Critical Instruction stack), which is CPI stack based on CPA. The main challenge in constructing CPCI stack is how to analyze a large number of paths because CPA often results in numerous critical paths. In this paper, we show that there are more than ten to the tenth power critical paths in the execution of only one thousand instructions in 35 benchmarks out of 48 from SPEC CPU2006. Then, we propose a statistical method to analyze all the critical paths and show a case study using the benchmarks.

Original languageEnglish
Pages (from-to)758-766
Number of pages9
JournalIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
VolumeE102A
Issue number6
DOIs
Publication statusPublished - Jan 1 2019

Fingerprint

Critical path analysis
Critical Path
Path Analysis
Cycle
Statistical methods
Path
Benchmark
Stacking
Statistical method
Execution Time
Counting
Optimise

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Computer Graphics and Computer-Aided Design
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

@article{a75a37a018ab40e29a20429c51a1d058,
title = "Critical path based microarchitectural bottleneck analysis for out-of-order execution",
abstract = "SUMMARY Correctly understanding microarchitectural bottlenecks is important to optimize performance and energy of OoO (Out-of-Order) processors. Although CPI (Cycles Per Instruction) stack has been utilized for this purpose, it stacks architectural events heuristically by counting how many times the events occur, and the order of stacking affects the result, which may be misleading. It is because CPI stack does not consider the execution path of dynamic instructions. Critical path analysis (CPA) is a well-known method to identify the critical execution path of dynamic instruction execution on OoO processors. The critical path consists of the sequence of events that determines the execution time of a program on a certain processor. We develop a novel representation of CPCI stack (Cycles Per Critical Instruction stack), which is CPI stack based on CPA. The main challenge in constructing CPCI stack is how to analyze a large number of paths because CPA often results in numerous critical paths. In this paper, we show that there are more than ten to the tenth power critical paths in the execution of only one thousand instructions in 35 benchmarks out of 48 from SPEC CPU2006. Then, we propose a statistical method to analyze all the critical paths and show a case study using the benchmarks.",
author = "Teruo Tanimoto and Takatsugu Ono and Inoue Koji",
year = "2019",
month = "1",
day = "1",
doi = "10.1587/transfun.E102.A.758",
language = "English",
volume = "E102A",
pages = "758--766",
journal = "IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences",
issn = "0916-8508",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "6",

}

TY - JOUR

T1 - Critical path based microarchitectural bottleneck analysis for out-of-order execution

AU - Tanimoto, Teruo

AU - Ono, Takatsugu

AU - Koji, Inoue

PY - 2019/1/1

Y1 - 2019/1/1

N2 - SUMMARY Correctly understanding microarchitectural bottlenecks is important to optimize performance and energy of OoO (Out-of-Order) processors. Although CPI (Cycles Per Instruction) stack has been utilized for this purpose, it stacks architectural events heuristically by counting how many times the events occur, and the order of stacking affects the result, which may be misleading. It is because CPI stack does not consider the execution path of dynamic instructions. Critical path analysis (CPA) is a well-known method to identify the critical execution path of dynamic instruction execution on OoO processors. The critical path consists of the sequence of events that determines the execution time of a program on a certain processor. We develop a novel representation of CPCI stack (Cycles Per Critical Instruction stack), which is CPI stack based on CPA. The main challenge in constructing CPCI stack is how to analyze a large number of paths because CPA often results in numerous critical paths. In this paper, we show that there are more than ten to the tenth power critical paths in the execution of only one thousand instructions in 35 benchmarks out of 48 from SPEC CPU2006. Then, we propose a statistical method to analyze all the critical paths and show a case study using the benchmarks.

AB - SUMMARY Correctly understanding microarchitectural bottlenecks is important to optimize performance and energy of OoO (Out-of-Order) processors. Although CPI (Cycles Per Instruction) stack has been utilized for this purpose, it stacks architectural events heuristically by counting how many times the events occur, and the order of stacking affects the result, which may be misleading. It is because CPI stack does not consider the execution path of dynamic instructions. Critical path analysis (CPA) is a well-known method to identify the critical execution path of dynamic instruction execution on OoO processors. The critical path consists of the sequence of events that determines the execution time of a program on a certain processor. We develop a novel representation of CPCI stack (Cycles Per Critical Instruction stack), which is CPI stack based on CPA. The main challenge in constructing CPCI stack is how to analyze a large number of paths because CPA often results in numerous critical paths. In this paper, we show that there are more than ten to the tenth power critical paths in the execution of only one thousand instructions in 35 benchmarks out of 48 from SPEC CPU2006. Then, we propose a statistical method to analyze all the critical paths and show a case study using the benchmarks.

UR - http://www.scopus.com/inward/record.url?scp=85069605493&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069605493&partnerID=8YFLogxK

U2 - 10.1587/transfun.E102.A.758

DO - 10.1587/transfun.E102.A.758

M3 - Article

VL - E102A

SP - 758

EP - 766

JO - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

JF - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

SN - 0916-8508

IS - 6

ER -