Journal ArticleDOI
Identifying the optimal level of parallelism in transactional memory applications
Reads0
Chats0
TLDR
A novel hybrid approach is introduced that combines model-driven performance forecasting techniques and on-line exploration in order to take the best of the two techniques, namely enhancing robustness despite model’s inaccuracies, and maximizing convergence speed towards optimum solutions.Abstract:
In this paper we investigate the issue of automatically identifying the "natural" degree of parallelism of an application using software transactional memory (STM), i.e., the workload-specific multiprogramming level that maximizes application's performance. We discuss the importance of adapting the concurrency level in two different scenarios, a shared-memory and a distributed STM infrastructure. We propose and evaluate two alternative self-tuning methodologies, explicitly tailored for the considered scenarios. In shared-memory STM, we show that lightweight, black-box approaches relying solely on on-line exploration can be extremely effective. For distributed STMs , we introduce a novel hybrid approach that combines model-driven performance forecasting techniques and on-line exploration in order to take the best of the two techniques, namely enhancing robustness despite model's inaccuracies, and maximizing convergence speed towards optimum solutions.read more
Citations
More filters
Proceedings Article
Self-Tuning Intel Transactional Synchronization Extensions
Nuno Diegues,Paolo Romano +1 more
TL;DR: This work introduces an innovative self-tuning approach that exploits lightweight reinforcement learning techniques to identify the optimal TSX configuration in a workload-oblivious manner, i.e. not requiring any off-line/a-priori sampling of the application’s workload.
Proceedings ArticleDOI
Enhancing Performance Prediction Robustness by Combining Analytical Modeling and Machine Learning
TL;DR: Several hybrid/gray box techniques are explored that exploit AM and ML in synergy in synergy to get the best of the two worlds, targeting two complex and widely adopted middleware systems.
Proceedings ArticleDOI
ProteusTM: Abstraction Meets Performance in Transactional Memory
TL;DR: The proposed system, ProteusTM, hides behind the TM interface a large library of implementations and leverages a novel multi-dimensional online optimization scheme, combining two popular learning techniques: Collaborative Filtering and Bayesian Optimization.
Proceedings ArticleDOI
A Combined Analytical Modeling Machine Learning Approach for Performance Prediction of MapReduce Jobs in Cloud Environment
TL;DR: This work proposes and validate a hybrid approach exploiting both queuing networks and support vector regression, in order to achieve a good accuracy without too many costly experiments on a real setup, and shows how the proposed approach attains a 21% improvement in accuracy over applying machine learning techniques without any support from analytical models.
References
More filters
Book
C4.5: Programs for Machine Learning
TL;DR: A complete guide to the C4.5 system as implemented in C for the UNIX environment, which starts from simple core learning methods and shows how they can be elaborated and extended to deal with typical problems such as missing data and over hitting.
Programs for Machine Learning
Steven L. Salzberg,Alberto Segre +1 more
TL;DR: In his new book, C4.5: Programs for Machine Learning, Quinlan has put together a definitive, much needed description of his complete system, including the latest developments, which will be a welcome addition to the library of many researchers and students.
Proceedings ArticleDOI
Transactional memory: architectural support for lock-free data structures
Maurice Herlihy,J. Eliot B. Moss +1 more
TL;DR: Simulation results show that transactional memory matches or outperforms the best known locking techniques for simple benchmarks, even in the absence of priority inversion, convoying, and deadlock.
Proceedings ArticleDOI
STAMP: Stanford Transactional Applications for Multi-Processing
TL;DR: This paper introduces the Stanford Transactional Application for Multi-Processing (STAMP), a comprehensive benchmark suite for evaluating TM systems and uses the suite to evaluate six different TM systems, identify their shortcomings, and motivate further research on their performance characteristics.
Related Papers (5)
Transactional memory: architectural support for lock-free data structures
Maurice Herlihy,J. Eliot B. Moss +1 more