2019 | OriginalPaper | Buchkapitel
GPUMixer: Performance-Driven Floating-Point Tuning for GPU Scientific Applications
verfasst von : Ignacio Laguna, Paul C. Wood, Ranvijay Singh, Saurabh Bagchi
Erschienen in: High Performance Computing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
GPUMixer
, a tool to perform mixed-precision floating-point tuning on scientific GPU applications. While precision tuning techniques are available, they are designed for serial programs and are accuracy-driven, i.e., they consider configurations that satisfy accuracy constraints, but these configurations may degrade performance. GPUMixer
, in contrast, presents a performance-driven approach for tuning. We introduce a novel static analysis that finds Fast Imprecise Sets (FISets), sets of operations on low precision that minimize type conversions, which often yield performance speedups. To estimate the relative error introduced by GPU mixed-precision, we propose shadow computations analysis for GPUs, the first of this class for multi-threaded applications. GPUMixer
obtains performance improvements of up to \(46.4\%\) of the ideal speedup in comparison to only \(20.7\%\) found by state-of-the-art methods.