A Machine Learning-Based Approach for Thread Mapping on Transactional Memory Applications - Archive ouverte HAL Access content directly
Conference Papers Year : 2011

A Machine Learning-Based Approach for Thread Mapping on Transactional Memory Applications

(1) , , (1) , , , (1)
1
Luis Fabricio Wanderley Góes
  • Function : Author
Christiane Pousa Ribeiro
  • Function : Author
  • PersonId : 856332
Murray Cole
  • Function : Author
Marcelo Cintra
  • Function : Author

Abstract

Thread mapping has been extensively used as a technique to efficiently exploit memory hierarchy on modern chip-multiprocessors. It places threads on cores in order to amortize memory latency and/or to reduce memory contention. However, efficient thread mapping relies upon matching application behavior with system characteristics. Particularly, Software Transactional Memory (STM) applications introduce another dimension due to its runtime system support. Existing STM systems implement several conflict detection and resolution mechanisms, which leads STM applications to behave differently for each combination of these mechanisms. In this paper we propose a machine learning-based approach to automatically infer a suitable thread mapping strategy for transactional memory applications. First, we profile several STM applications from the STAMP benchmark suite considering application, STM system and platform features to build a set of input instances. Then, such data feeds a machine learning algorithm, which produces a decision tree able to predict the most suitable thread mapping strategy for new unobserved instances. Results show that our approach improves performance up to 18.46% compared to the worst case and up to 6.37% over the Linux default thread mapping strategy.
Not file

Dates and versions

hal-00788791 , version 1 (15-02-2013)

Identifiers

Cite

Marcio Bastos Castro, Luis Fabricio Wanderley Góes, Christiane Pousa Ribeiro, Murray Cole, Marcelo Cintra, et al.. A Machine Learning-Based Approach for Thread Mapping on Transactional Memory Applications. 18th Annual and International Conference on High Performance Computing (HiPC), 2011, Bangalore, India. pp.1-10, ⟨10.1109/HiPC.2011.6152736⟩. ⟨hal-00788791⟩
160 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More