PARSECSs
Citations Over TimeTop 10% of 2015 papers
Abstract
In this work, we show how parallel applications can be implemented efficiently using task parallelism. We also evaluate the benefits of such parallel paradigm with respect to other approaches. We use the PARSEC benchmark suite as our test bed, which includes applications representative of a wide range of domains from HPC to desktop and server applications. We adopt different parallelization techniques, tailored to the needs of each application, to fully exploit the task-based model. Our evaluation shows that task parallelism achieves better performance than thread-based parallelization models, such as Pthreads. Our experimental results show that we can obtain scalability improvements up to 42% on a 16-core system and code size reductions up to 81%. Such reductions are achieved by removing from the source code application specific schedulers or thread pooling systems and transferring these responsibilities to the runtime system software.
Related Papers
- → TP-PARSEC: A Task Parallel PARSEC Benchmark Suite(2019)3 cited
- → Nested parallelism concepts of ray tracing algorithms and multithreading API performance analysis(2016)1 cited
- → Sthreads: Multithreading for SCI clusters(1999)1 cited
- → Multi-lingual threading(2002)
- A parallel formulation for the simulation of a generic branch predictor(2012)