LIKWID: Lightweight Performance Tools

TL;DR

LIKWID is a lightweight set of command line tools designed to help optimize microprocessor performance by providing topology probing, thread affinity enforcement, performance measurement, and microbenchmarking, especially for shared-memory and MPI applications.

Contribution

It introduces a comprehensive tool set that simplifies performance analysis and optimization of microarchitectures, including portable thread affinity and hardware counter integration.

Findings

Demonstrated the impact of thread affinity on OpenMP benchmark performance.

Used hardware counters to analyze stencil code performance.

Showed how to detect bandwidth issues on ccNUMA nodes.

Abstract

Exploiting the performance of today's microprocessors requires intimate knowledge of the microarchitecture as well as an awareness of the ever-growing complexity in thread and cache topology. LIKWID is a set of command line utilities that addresses four key problems: Probing the thread and cache topology of a shared-memory node, enforcing thread-core affinity on a program, measuring performance counter metrics, and microbenchmarking for reliable upper performance bounds. Moreover, it includes a mpirun wrapper allowing for portable thread-core affinity in MPI and hybrid MPI/threaded applications. To demonstrate the capabilities of the tool set we show the influence of thread affinity on performance using the well-known OpenMP STREAM triad benchmark, use hardware counter tools to study the performance of a stencil code, and finally show how to detect bandwidth problems on ccNUMA-based compute nodes.