Estimating CPI Stacks From Multiplexed Performance Counter Data Using Machine Learning
Optimizing software at runtime is much easier with a clear understanding of the bottlenecks facing the software. CPI stacks are a common method of visualizing these bottlenecks. However, existing proposals to implement CPI stacks require hardware modifications. To compute CPI stacks without modifying the CPU, we demonstrate CPI stacks can be estimated from existing performance counters using machine learning.
Authors
- Jered Dominguez-Trujillo (ORCID: https://orcid.org/0000-0001-7021-3077)
- Galen Shipman (ORCID: https://orcid.org/0000-0001-6297-2145)
- Kevin J. Sheridan
- Paul V. Gratz (ORCID: https://orcid.org/0000-0001-7120-7189)
- Jiang Hu (ORCID: https://orcid.org/0000-0003-1157-7799)
- Daniel Puckett
- Tyler Tomer
Institutions
- Los Alamos National Laboratory (US)
- Texas A&M University (US)
Publication Details
- Journal
- IEEE Computer Architecture Letters
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1109/lca.2025.3556644
- Citations
- 1
- Primary Topic
- Software System Performance and Reliability
- Type
- article
- Field-Weighted Citation Impact
- 1.49