Experiences using adaptive concurrency in transactional memory with Lee's routing algorithm

  • Authors:
  • Mohammad Ansari;Christos Kotselidis;Kim Jarvis;Mikel Lujan;Chris Kirkham;Ian Watson

  • Affiliations:
  • University of Manchester, Manchester, United Kngdm;University of Manchester, Manchester, United Kngdm;University of Manchester, Manchester, United Kngdm;University of Manchester, Manchester, United Kngdm;University of Manchester, Manchester, United Kngdm;University of Manchester, Manchester, United Kngdm

  • Venue:
  • Proceedings of the 13th ACM SIGPLAN Symposium on Principles and practice of parallel programming
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Experience in profiling Lee's routing algorithm, a new complex TM application, showed that transactional applications may exhibit dynamic exploitable parallelism, i.e. the amount of useful parallelism available at any point in time varies during the execution of the application. Obviously, executing too many transactions at times when the available parallelism is low will lead to high contention and wasted computation in aborted transactions, and vice versa. Current Transactional Memory (TM) implementations do not account for this behavior. This work employs adaptive concurrency to dynamically adjust the number of threads executing transactions concurrently. Our preliminary evaluation is performed in DSTM2 using Lee's routing algorithm, both of which were simple to modify to enable adaptive concurrency, and shows significant reduction in resource usage, and modest performance gains.