Abstract:

We investigate to which extent data partitioning can help improve the performance of software transactional memory (STM). Our main idea is that the access patterns of the various data structures of an application might be sufficiently different so that it would be beneficial to tune the behavior of the STM for individual data partitions. We evaluate our approach using standard transactional memory benchmarks. We show that these applications contain partitions with different characteristics and, despite the runtime overhead introduced by partition tracking and dynamic tuning, that partitioning provides significant performance improvements.

Bibtex:

@inproceedings{Riegel2008partitioning,
  author = {{T}orvald {R}iegel and {C}hristof {F}etzer and {P}ascal {F}elber},
  title = {{A}utomatic {D}ata {P}artitioning in {S}oftware {T}ransactional {M}emories},
  booktitle = {20th ACM Symposium on Parallelism in Algorithms and Architectures (SPAA)},
  year = {2008},
}

Download:

• Automatic Data Partitioning in Software Transactional Memories (PDF)
• Page with links to PDF and BibTex files
• Source code for the compiler tool (LLVM-based) and the STM runtime can be downloaded at tinystm.org