Using ChatGPT on Improving Program Performance with pprof and Benchmark

Wei Cheng Lei; Luo You Jian; Yan Wen Chen; Li Der Chou

doi:10.1109/ICCCI59363.2023.10210148

Using ChatGPT on Improving Program Performance with pprof and Benchmark

Wei Cheng Lei, Luo You Jian, Yan Wen Chen, Li Der Chou

Department of Computer Science and Information Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In the context of limited computing resources, optimizing program architecture is crucial. Therefore, this paper proposes to apply the powerful analytical capabilities of large language models (LLM) to the field of systematic performance optimization. The output of pprof and benchmark is fed into ChatGPT, and the program's performance is improved based on feedback. In the case study, the number of memory allocations for the objective function was successfully reduced from 99 to 1, resulting in a reduction of the test execution time from 8.6 microseconds to 0.36 microseconds. At the same time, the memory allocation was also reduced from 53.5KB to approximately 1KB.

Original language	English
Title of host publication	2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	256-260
Number of pages	5
ISBN (Electronic)	9798350326956
DOIs	https://doi.org/10.1109/ICCCI59363.2023.10210148
State	Published - 2023
Event	5th International Conference on Computer Communication and the Internet, ICCCI 2023 - Fujisawa, Japan Duration: 23 Jun 2023 → 25 Jun 2023

Publication series

Name	2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023

Conference

Conference	5th International Conference on Computer Communication and the Internet, ICCCI 2023
Country/Territory	Japan
City	Fujisawa
Period	23/06/23 → 25/06/23

Keywords

Large Language Models
Performance analysis
Prompt Engineering
System optimization

Access to Document

10.1109/ICCCI59363.2023.10210148

基於eBPF微服務整合管理框架之設計與實作(1/2)
Chou, L.-D. (PI)
1/08/22 → 31/07/23
Project: Research
基於AI之5G/B5G網路切片管理與安全技術之研究(1/3)
Chou, L.-D. (PI)
1/08/22 → 31/07/23
Project: Research

Cite this

Lei, W. C., Jian, L. Y., Chen, Y. W., & Chou, L. D. (2023). Using ChatGPT on Improving Program Performance with pprof and Benchmark. In 2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023 (pp. 256-260). (2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCCI59363.2023.10210148

Lei, Wei Cheng ; Jian, Luo You ; Chen, Yan Wen et al. / Using ChatGPT on Improving Program Performance with pprof and Benchmark. 2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 256-260 (2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023).

@inproceedings{6d2ce769a6414b819020e5b15421f7f6,

title = "Using ChatGPT on Improving Program Performance with pprof and Benchmark",

abstract = "In the context of limited computing resources, optimizing program architecture is crucial. Therefore, this paper proposes to apply the powerful analytical capabilities of large language models (LLM) to the field of systematic performance optimization. The output of pprof and benchmark is fed into ChatGPT, and the program's performance is improved based on feedback. In the case study, the number of memory allocations for the objective function was successfully reduced from 99 to 1, resulting in a reduction of the test execution time from 8.6 microseconds to 0.36 microseconds. At the same time, the memory allocation was also reduced from 53.5KB to approximately 1KB.",

keywords = "Large Language Models, Performance analysis, Prompt Engineering, System optimization",

author = "Lei, {Wei Cheng} and Jian, {Luo You} and Chen, {Yan Wen} and Chou, {Li Der}",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 5th International Conference on Computer Communication and the Internet, ICCCI 2023 ; Conference date: 23-06-2023 Through 25-06-2023",

year = "2023",

doi = "10.1109/ICCCI59363.2023.10210148",

language = "???core.languages.en_GB???",

series = "2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "256--260",

booktitle = "2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023",

}

Lei, WC, Jian, LY, Chen, YW & Chou, LD 2023, Using ChatGPT on Improving Program Performance with pprof and Benchmark. in 2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023. 2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023, Institute of Electrical and Electronics Engineers Inc., pp. 256-260, 5th International Conference on Computer Communication and the Internet, ICCCI 2023, Fujisawa, Japan, 23/06/23. https://doi.org/10.1109/ICCCI59363.2023.10210148

Using ChatGPT on Improving Program Performance with pprof and Benchmark. / Lei, Wei Cheng; Jian, Luo You; Chen, Yan Wen et al.
2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 256-260 (2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Using ChatGPT on Improving Program Performance with pprof and Benchmark

AU - Lei, Wei Cheng

AU - Jian, Luo You

AU - Chen, Yan Wen

AU - Chou, Li Der

PY - 2023

Y1 - 2023

N2 - In the context of limited computing resources, optimizing program architecture is crucial. Therefore, this paper proposes to apply the powerful analytical capabilities of large language models (LLM) to the field of systematic performance optimization. The output of pprof and benchmark is fed into ChatGPT, and the program's performance is improved based on feedback. In the case study, the number of memory allocations for the objective function was successfully reduced from 99 to 1, resulting in a reduction of the test execution time from 8.6 microseconds to 0.36 microseconds. At the same time, the memory allocation was also reduced from 53.5KB to approximately 1KB.

AB - In the context of limited computing resources, optimizing program architecture is crucial. Therefore, this paper proposes to apply the powerful analytical capabilities of large language models (LLM) to the field of systematic performance optimization. The output of pprof and benchmark is fed into ChatGPT, and the program's performance is improved based on feedback. In the case study, the number of memory allocations for the objective function was successfully reduced from 99 to 1, resulting in a reduction of the test execution time from 8.6 microseconds to 0.36 microseconds. At the same time, the memory allocation was also reduced from 53.5KB to approximately 1KB.

KW - Large Language Models

KW - Performance analysis

KW - Prompt Engineering

KW - System optimization

UR - http://www.scopus.com/inward/record.url?scp=85170028267&partnerID=8YFLogxK

U2 - 10.1109/ICCCI59363.2023.10210148

DO - 10.1109/ICCCI59363.2023.10210148

M3 - 會議論文篇章

AN - SCOPUS:85170028267

T3 - 2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023

SP - 256

EP - 260

BT - 2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 5th International Conference on Computer Communication and the Internet, ICCCI 2023

Y2 - 23 June 2023 through 25 June 2023

ER -

Lei WC, Jian LY, Chen YW, Chou LD. Using ChatGPT on Improving Program Performance with pprof and Benchmark. In 2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 256-260. (2023 5th International Conference on Computer Communication and the Internet, ICCCI 2023). doi: 10.1109/ICCCI59363.2023.10210148

Using ChatGPT on Improving Program Performance with pprof and Benchmark

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Projects

基於eBPF微服務整合管理框架之設計與實作(1/2)

基於AI之5G/B5G網路切片管理與安全技術之研究(1/3)

Cite this