Transit: A visual analytical model for multithreaded machines
Publikation: Beitrag in Buch/Konferenzbericht/Sammelband/Gutachten › Beitrag in Konferenzband › Beigetragen › Begutachtung
Beitragende
Abstract
With the extraordinary growth of cores and threads in to-day's multithreaded machines, analyzing and tuning the performance of such platforms becomes a challenging task. In this paper, we propose an intuitive and visualizable model for analyzing the performance of contemporary highly con-current multithreaded machines. Based on ow balancing between service demand and service supply of the memory system, the model draws an intuitive figure to characterize machine state, identify bottlenecks and determine optimization directions. The tractability of the model is highlighted as it only requires two parameters from the workload. Our model achieves 90% and 83% prediction accuracy for computation throughput on Fermi and Kepler GPUs over the 16 applications from Rodinia benchmark.
Details
Originalsprache | Englisch |
---|---|
Titel | HPDC 2015 - Proceedings of the 24th International Symposium on High-Performance Parallel and Distributed Computing |
Herausgeber (Verlag) | Association for Computing Machinery, Inc |
Seiten | 101-106 |
Seitenumfang | 6 |
ISBN (elektronisch) | 9781450335508 |
Publikationsstatus | Veröffentlicht - 15 Juni 2015 |
Peer-Review-Status | Ja |
Extern publiziert | Ja |
Konferenz
Titel | 24th ACM Symposium on High-Performance Parallel and Distributed Computing, HPDC 2015 |
---|---|
Dauer | 15 - 19 Juni 2015 |
Stadt | Portland |
Land | USA/Vereinigte Staaten |
Schlagworte
Forschungsprofillinien der TU Dresden
ASJC Scopus Sachgebiete
Schlagwörter
- GPUs, Multithreaded machine, Performance modeling, Performance optimization