A Fast Causal Profiler for Task Parallel Programs

Adarsh Yoga; Santosh Nagarakatte

A Fast Causal Profiler for Task Parallel Programs

Programming Languages 2017-07-04 v3 Distributed, Parallel, and Cluster Computing

Authors: Adarsh Yoga , Santosh Nagarakatte

Abstract

This paper proposes TASKPROF, a profiler that identifies parallelism bottlenecks in task parallel programs. It leverages the structure of a task parallel execution to perform fine-grained attribution of work to various parts of the program. TASKPROF's use of hardware performance counters to perform fine-grained measurements minimizes perturbation. TASKPROF's profile execution runs in parallel using multi-cores. TASKPROF's causal profile enables users to estimate improvements in parallelism when a region of code is optimized even when concrete optimizations are not yet known. We have used TASKPROF to isolate parallelism bottlenecks in twenty three applications that use the Intel Threading Building Blocks library. We have designed parallelization techniques in five applications to in- crease parallelism by an order of magnitude using TASKPROF. Our user study indicates that developers are able to isolate performance bottlenecks with ease using TASKPROF.

Keywords

parallel programming parallel algorithm flaky test detection

Cite

@article{arxiv.1705.01522,
  title  = {A Fast Causal Profiler for Task Parallel Programs},
  author = {Adarsh Yoga and Santosh Nagarakatte},
  journal= {arXiv preprint arXiv:1705.01522},
  year   = {2017}
}

Comments

11 pages

A Fast Causal Profiler for Task Parallel Programs

Abstract

Keywords

Cite

Comments

Related papers