Logo des Repositoriums
 
Zeitschriftenartikel

Towards Predicting Performance of GPU-dependent Applications on the Example of Machine Learning in Enterprise Applications

Zusammenfassung

Algorithms processed by Graphics Processing Units (GPU) became popular recently. Bitcoin mining algorithms, image processing and all types of machine learning are famous examples for that. Infrastructureas-a-Service provider picked up this trend and offer graphics processing power as part of their service portfolio. The performance gains when choosing a GPU implementation can be enormous. Designing and implementing a GPU-depended algorithm has some fundamental differences compared to classical algorithms, but not all algorithmic problems benefit from GPU usage regarding the overall performance and response time. Especially the interaction between Central Processing Unit (CPU) and GPU must be considered as it can become a bottleneck. Predicting and comparing the performance of GPU-depended applications in combination with their corresponding CPUs allows to assist design decisions in modern applications. In this work, we present concepts on how to predict algorithm performance relying on GPU processing and their relationship with the CPU using the Palladio Component Model and the Palladio Bench.

Beschreibung

Willnecker, Felix; Krcmar, Helmut (2017): Towards Predicting Performance of GPU-dependent Applications on the Example of Machine Learning in Enterprise Applications. Softwaretechnik-Trends Band 37, Heft 3. Bonn: Geselllschaft für Informatik e.V.. PISSN: 0720-8928. pp. 32-34. Sonderteil: Proceedings of the 8th Symposium on Software Performance (SSP), Karlsruhe, 09. - 10. November 2017

Schlagwörter

Zitierform

DOI

Tags