PIRA: Performance Instrumentation Refinement Automation

" summary

PIRA is an automation tool that refines Score-P profiling to focus on relevant program regions by combining static source-code features and dynamic runtime information, using the whole-program call-graph framework MetaCG. The system performs automatic hotspot detection and refinement, scalability analysis, kernel identification, and MPI load-imbalance detection; experiments on SPEC CPU 2006, LULESH, and ISSM demonstrate overview measurements with <10–15% runtime overhead. Both PIRA and MetaCG are open-source under BSD 3-clause on GitHub.

AI-written summary · about

" related talks

• 24mPythia: An Automated, Cross-layer Instrumentation Framework for Diagnosing Performanc...Emre Ates · Observability Practitioners Summit 2019
• 16mPilot Execution: Simulating Failure Recovery In Situ for Production Distributed SystemsZhenyu Li, Angting Cai, Chang Lou · NSDI '26
• 17mEROICA: Online Performance Troubleshooting for Large-scale Model TrainingYang Zhang, Yong Li, Dennis Cai, Ennan Zhai, Tianyin Xu, Yu Guan, Zhiyu Yin, Haoyu Chen, Sheng Cheng, Chaojie Yang, Kun Qian, Pengcheng Zhang, Hanyu Zhao · NSDI '26
• 15mCollie: Finding Performance Anomalies in RDMA SubsystemsZhuo Jiang, Jianxi Ye, Chuanxiong Guo, Xinhao Kong, Danyang Zhuo, Yibo Zhu, Huaping Zhou · 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI '22)
• 12mFlightplan: Dataplane Disaggregation and Placement for P4 ProgramsBoon Thau Loo, Isaac Pedisich, John Sonchack, Nik Sultana, Hans Giesen, Zhaoyang Han, Nishanth Shyamkumar, Shivani Burad, André DeHon · NSDI ’21
• 24mPLOVER: Fast, Multi-core Scalable Virtual Machine Fault-toleranceCheng Wang · NSDI '18
• 01h 00mSystems @Scale Summer Q&A Week 4Yun Jin · Systems @Scale Summer 2021
• 29mScaleCheck: A Single-Machine Approach for Discovering Scalability Bugs...Cesar A. Stuardo · FAST '19