Effective Performance Issue Diagnosis with Value-Assisted Cost Profiling
Lingmei Weng, Yigong Hu, Peng Huang, Jason Nieh, Junfeng Yang
摘要
Diagnosing performance issues is often difficult, especially when they occur only during some program executions. Profilers can help with performance debugging, but are ineffective when the most costly functions are not the root causes of performance issues. To address this problem, we introduce a new profiling methodology, value-assisted cost profiling, and a tool vProf. Our insight is that capturing the values of variables can greatly help diagnose performance issues. vProf continuously records values while profiling normal and buggy program executions. It identifies anomalies in the values and the functions where they occur to pinpoint the real root causes of performance issues. Using a set of 15 real-world performance bugs in four widely used applications, we show that vProf is effective at diagnosing all of the issues while other stateof-the-art tools diagnose only a few of them. We further use vProf to diagnose longstanding performance issues in these applications that have been unresolved for over four years.
• Software and its engineering → Software testing and debugging.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了最后一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper5
- FastPERT: Towards Fast Microservice Application Latency Prediction via Structural Inductive Bias over PERT NetworksDa Sun Handason Tam, Huanle Xu, Yang Liu, Siyue Xie 等AAAI 2025 · 被引用 5 次
- Pushing Performance Isolation Boundaries into Application with pBoxYigong Hu, Gongqi Huang, Peng HuangSOSP 2023 · 被引用 2 次
- More Than Just Functional: LLM-as-a-Critique for Efficient Code GenerationDerui Zhu, Dingfan Chen, Jinfu Chen, Jens Grossklags 等NeurIPS 2025 · 被引用 2 次
- Diagnosing Performance Issues in Application-Defined ResourcesYigong Hu, You-Liang Huang, Haodong Zheng, Yicheng Liu 等OSDI 2026 · 被引用 1 次
- Mitigating Application Resource Overload with Targeted Task CancellationYigong Hu, Zeyin Zhang, Yicheng Liu, Yile Gu 等SOSP 2025
它引用的顶会 Paper5
- Automated Reasoning and Detection of Specious Configuration in Large Systems with Symbolic ExecutionYigong Hu, Gongqi Huang, Peng HuangOSDI 2020 · 被引用 31 次
- DMon: Efficient Detection and Correction of Data Locality Problems Using Selective ProfilingTanvir Ahmed Khan, Ian Neal, Gilles Pokam, Barzan Mozafari 等OSDI 2021 · 被引用 31 次
- On Debugging the Performance of Configurable Software Systems: Developer Needs and Tailored Tool SupportMiguel Velez, Pooyan Jamshidi, Norbert Siegmund, Sven Apel 等ICSE 2022 · 被引用 21 次
- Argus: Debugging Performance Issues in Modern Desktop Applications with Annotated Causal TracingLingmei Weng, Peng Huang, Jason Nieh, Junfeng YangUSENIX ATC 2021 · 被引用 16 次
- Analyzing system performance with probabilistic performance annotationsDaniele Rogora, Antonio Carzaniga, Amer Diwan, Matthias Hauswirth 等EuroSys 2020 · 被引用 11 次
相关 Paper
- Vapro: performance variance detection and diagnosis for production-run parallel applicationsLiyan Zheng, Jidong Zhai, Xiongchao Tang, Haojie Wang 等PPoPP 2022 · 被引用 15 次
- LDB: An Efficient Latency Profiling Tool for Multithreaded ApplicationsInho Cho, Seo Jin Park, Ahmed Saeed, Mohammad Alizadeh 等NSDI 2024 · 被引用 4 次
- Relational Debugging - Pinpointing Root Causes of Performance ProblemsXiang (Jenny) Ren, Sitao Wang, Zhuqi Jin, David Lion 等OSDI 2023 · 被引用 5 次
- DiagConfig: Configuration Diagnosis of Performance Violations in Configurable Software SystemsZhiming Chen, Pengfei Chen, Peipei Wang, Guangba Yu 等FSE 2023 · 被引用 9 次
- ValueExpert: exploring value patterns in GPU-accelerated applicationsKeren Zhou, Yueming Hao, John M. Mellor-Crummey, Xiaozhu Meng 等ASPLOS 2022 · 被引用 19 次
