A quantitative study of irregular programs on GPUs

By Martin Burtscher, Rupesh Nasre, Keshav Pingali. IEEE International Symposium on Workload Characterization, 2012

397 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.

Zorua: A holistic approach to resource virtualization in GPUs (2016, methodology): “Within this window, the GPU can support two thread blocks running concurrently for MST.”
Accel-Sim: An Extensible Simulation Framework for Validated GPU Modeling (2018, methodology): “oads. We indiscriminately include all the workloads we could get to run in simulation from the original GPGPU-Sim paper [8], Rodinia 3.1 [22], the CUDA SDK [23], Parboil [24], Pannotia [25], Lonestar [26], SHOC [27], and a collection of DOE proxy apps [28]. The sum total of which is over 1400 unique”
MASK: Redesigning the GPU Memory Hierarchy to Support Multi-Application Concurrency (2018, background): “they provide the highest-available compute density and remain the platform of choice in many domains [1, 11, 28, 31, 45].”
Gunrock: a high-performance graph processing library on the GPU (2015, background): “Efficiently implementing SSSP continues to be an interesting problem in the GPU world [3, 5, 6].”