Modeling transactional memory workload performance

Transactional memory promises to make parallel programming easier than with fine-grained locking, while performing just as well. This performance claim is not always borne out because an applica-tion may violate a common-case assumption of the TM designer or because of external system effects. In order to help programmers assess the suitability of their code for transactional memory, this work introduces a formal model of transactional memory as well as a tool, called Syncchar. Syncchar can predict the speedup of a conversion from locks to transactions within 25 % for the STAMP benchmarks. Because getting good performance from transactions is more difficult than commonly appreciated, developers need tools to tune transactional performance. Categories and Subject Descriptors C.1.4 [Processor Architec-tures]: [Parallel Architecture]; C.4 [Processor Architectures]: Per-formance of Systems—Modeling Technique