The PARSEC benchmark suite

TLDR

Previous benchmarks for multiprocessors focused on high‑performance computing applications and used a limited set of synchronization methods. This paper introduces and characterizes the Princeton Application Repository for Shared‑Memory Computers (PARSEC), a benchmark suite for studying Chip‑Multiprocessors (CMPs). PARSEC comprises emerging recognition, mining, and synthesis applications along with system workloads that emulate large‑scale multithreaded commercial programs. Characterization shows that PARSEC spans diverse working sets, locality, data sharing, synchronization, and off‑chip traffic patterns, and it is publicly available.

Abstract

This paper presents and characterizes the Princeton Application Repository for Shared-Memory Computers (PARSEC), a benchmark suite for studies of Chip-Multiprocessors (CMPs). Previous available benchmarks for multiprocessors have focused on high-performance computing applications and used a limited number of synchronization methods. PARSEC includes emerging applications in recognition, mining and synthesis (RMS) as well as systems applications which mimic large-scale multithreaded commercial programs. Our characterization shows that the benchmark suite covers a wide spectrum of working sets, locality, data sharing, synchronization and off-chip traffic. The benchmark suite has been made available to the public.