RZBENCH: Performance evaluation of current HPC architectures using low-level and application benchmarks

TL;DR

RZBENCH is a comprehensive benchmark suite designed to evaluate the performance of current HPC architectures with application and low-level benchmarks, emphasizing maintainability and relevance for scientific supercomputing.

Contribution

The paper introduces RZBENCH, a tailored benchmark suite for HPC systems, and provides performance data across diverse architectures, highlighting the need for critical assessment of benchmark results.

Findings

Performance data for various architectures including HLRB-II and InfiniBand clusters.

Comparison of benchmark results across different system architectures.

Emphasis on the importance of critical review of performance benchmarks.

Abstract

RZBENCH is a benchmark suite that was specifically developed to reflect the requirements of scientific supercomputer users at the University of Erlangen-Nuremberg (FAU). It comprises a number of application and low-level codes under a common build infrastructure that fosters maintainability and expandability. This paper reviews the structure of the suite and briefly introduces the most relevant benchmarks. In addition, some widely known standard benchmark codes are reviewed in order to emphasize the need for a critical review of often-cited performance results. Benchmark data is presented for the HLRB-II at LRZ Munich and a local InfiniBand Woodcrest cluster as well as two uncommon system architectures: A bandwidth-optimized InfiniBand cluster based on single socket nodes ("Port Townsend") and an early version of Sun's highly threaded T2 architecture ("Niagara 2").