Program optimization space pruning for a multithreaded gpu

By Shane Ryoo, Christopher I. Rodrigues, Sam S. Stone, Sara S. Baghsorkhi, Sain-Zee Ueng, John A. Stratton, Wen-mei W. Hwu. IEEE/ACM International Symposium on Code Generation and Optimization, 2008

316 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

OpenMP to GPGPU: a compiler framework for automatic translation and optimization (2009, methodology): “Several studies have been conducted to optimize the performance of CUDA-based GPGPU applications; an optimization space pruning technique [15] has been proposed, using a Pareto-optimal curve, to find the optimal configuration for a GPGPU application.”
Polyhedral parallel code generation for CUDA (2013, methodology): “have shown that the optimization space for CUDA programs can be highly non-linear, and highly challenging for model-based heuristics, performance modeling, or feedback-directed methods [Ryoo et al. 2008b; Ryoo et al. 2008a].”
Zorua: A holistic approach to resource virtualization in GPUs (2016, methodology): “For a naive programmer, this is a challenging task [14, 47, 59, 60, 61, 65] because, in addition to selecting a specification suited to an algorithm, the programmer needs to be aware of the details of the GPU architecture to fit the specification to the underlying hardware resources.”
An analytical model for a GPU architecture with memory-level and thread-level parallelism awareness (2009, background): “[24] proposed two metrics to reduce optimization spaces for programmers by calculating utilization and efficiency of applications.”
Analyzing CUDA workloads using a detailed GPU simulator (2009, background): “Thread scheduling inside a shader core is performed with zero overhead on a ﬁne-grained basis.”