Brief announcement: selfishness in transactional memory

Authors:
Raphael P. Eidenbenz;Roger Wattenhofer
Affiliations:
ETH Zurich, Zurich, Switzerland;ETH Zurich, Zurich, Switzerland
Venue:
Proceedings of the twenty-first annual symposium on Parallelism in algorithms and architectures
Year:
2009

Citing 5
Cited 1

Transactional memory: architectural support for lock-free data structures

ISCA '93 Proceedings of the 20th annual international symposium on computer architecture
Advanced contention management for dynamic software transactional memory

Proceedings of the twenty-fourth annual ACM symposium on Principles of distributed computing
Toward a theory of transactional contention managers

Proceedings of the twenty-fourth annual ACM symposium on Principles of distributed computing
Transactional contention management as a non-clairvoyant scheduling problem

Proceedings of the twenty-fifth annual ACM symposium on Principles of distributed computing
A flexible framework for implementing software transactional memory

Proceedings of the 21st annual ACM SIGPLAN conference on Object-oriented programming systems, languages, and applications

Good Programming in Transactional Memory

ISAAC '09 Proceedings of the 20th International Symposium on Algorithms and Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

In order to be efficient with selfish programmers, a multicore transactional memory (TM) system must be designed such that it is compatible with good programming incentives (GPI), i.e., writing efficient code for the overall system coincides with writing code that optimizes an individual program's performance. By implementing a selfish strategy, we show that under most contention managers (CM) proposed in the literature so far, TM systems are not GPI compatible, whereas a simple randomized CM is GPI compatible.