2021•arXiv (Cornell University)Open access

A Precise Program Phase Identification Method Based on Frequency Domain Analysis

Hsuan-Yi Lin, Ren‐Song Tsay

Open full text 0 citations

Abstract

In this paper, we present a systematic approach that transforms the program execution trace into the frequency domain and precisely identifies program phases. The analyzed results can be embedded into program code to mark the starting point and execution characteristics, such as CPI (Cycles per Instruction), of each phase. The so generated information can be applied to runtime program phase prediction. With the precise program phase information, more intelligent software and system optimization techniques can be further explored and developed.

Open-access reader

About this research paper

What this paper is about

In this paper, we present a systematic approach that transforms the program execution trace into the frequency domain and precisely identifies program phases. The analyzed results can be embedded into program code to mark the starting point and execution characteristics, such as CPI (Cycles per Instruction), of each phase. The so generated information can be applied to runtime program phase prediction. With the precise program phase information, more intelligent software and system optimization techniques can be further explored and developed.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

In this paper, we present a systematic approach that transforms the program execution trace into the frequency domain and precisely identifies program phases. The analyzed results can be embedded into program code to mark the starting point and execution characteristics, such as CPI (Cycles per Instruction), of each phase. The so generated information can be applied to runtime program phase prediction. With the precise program phase information, more intelligent software and system optimization techniques can be further explored and developed.

Key concepts: Identification (biology), Frequency domain, Computer science, Phase (matter), Domain (mathematical analysis), Mathematics, Physics, Mathematical analysis

Related papers

Back to paper searchBrowse research topicsOriginal source
A Precise Program Phase Identification Method Based on Frequency Domain Analysis — Research Paper | ScholarLens