Toward a Standard Benchmark Format and Suite for Floating-Point Analysis

Abstract

We introduce FPBench, a standard benchmark format for validation and optimization of numerical accuracy in floating-point computations. FPBench is a first step toward addressing an increasing need in our community for comparisons and combinations of tools from different application domains. To this end, FPBench provides a basic floatingpoint benchmark format and accuracy measures for comparing different tools. The FPBench format and measures allow comparing and composing different floating-point tools. We describe the FPBench format and measures and show that FPBench expresses benchmarks from recent papers in the literature, by building an initial benchmark suite drawn from these papers. We intend for FPBench to grow into a standard benchmark suite for the members of the floating-point tools research community. 3 We are using |S| to denote the number of elements in a set S 4 Where n is the number of arguments. 5 Where N is the number of valid inputs, and n is the number of arguments.

Finding root causes of floating point error

“…The first subsection demonstrates this for Herbgrind in its default configuration, while the second subsection examines the effect of Herbgrind's various tunable parameters. Each experiment uses the standard FPBench suite of general-purpose floating-point programs [10].…”

A Probabilistic Approach to Floating-Point Arithmetic

“…Analytically, the probability of overflow is 0.0775%, and empirically 0.0642% of the samples overflow. Next, we consider two benchmarks from FPBench [3] which our current setup can handle. In both cases we take halfprecision as the working precision.…”

Daisy - Framework for Analysis and Optimization of Numerical Programs (Tool Paper)

“…To make it userfriendly, we adopt the input format of Rosa, which is a real-valued functional domain-specific language in Scala. Unlike other tools today, which have custom input formats [41] or use prefix notation [12], Daisy's input is easily human readable 1 and natural to use. Daisy is itself written in the Scala programming language [35] and has limited and optional dependencies, making it portable and easy to install.…”