Performance variation across benchmark suites

Publisher Website
Google Scholar
Abstract
The performance ratio between two systems tends to vary across different benchmarks. Here we study this variation as a "signature" or "fingerprint" of the systems under consideration. This "fingerprint" can be used to guess the performance of programs not represented in a benchmark suite, assess the breadth and credibility of the benchmark suite, and infer details of the system design.

This publication has 4 references indexed in Scilit: