2017 · 70 citations · 25 references
ComputingComputer EngineeringComputer ScienceComputer GraphicsOperating SystemsParallel ComputingCompute Resource ManagementGraphics Processing Unit ArchitectureGraphics Processing Unit ComputingGraphics Processing UnitsHardware AccelerationKernel DesignDynamic Resource AllocationDynamic Resource ManagementGpu AcceleratorsGpu ArchitectureGpu Runtime SystemsKernel OptimizationMultitaskingStreaming Multiprocessor Design
As graphics processing units (GPUs) are broadly adopted, running multiple applications on a GPU at the same time is beginning to attract wide attention. Recent proposals on multitasking GPUs have focused on either spatial multitasking, which partitions GPU resource at a streaming multiprocessor (SM) granularity, or simultaneous multikernel (SMK), which runs multiple kernels on the same SM. However, multitasking performance varies heavily depending on the resource partitions within each scheme, and the application mixes. In this paper, we propose GPU Maestro that performs dynamic resource management for efficient utilization of multitasking GPUs. GPU Maestro can discover the best performing GPU resource partition exploiting both spatial multitasking and SMK. Furthermore, dynamism within a kernel and interference between the kernels are automatically considered because GPU Maestro finds the best performing partition through direct measurements. Evaluations show that GPU Maestro can improve average system throughput by 20.2% and 13.9% over the baseline spatial multitasking and SMK, respectively.
25
Rodinia: A benchmark suite for heterogeneous computing
Shuai Che, Michael Boyer, Jiayuan Meng et al. · 2009 · 3K citations
Analyzing CUDA workloads using a detailed GPU simulator
Ali Bakhoda, George L. Yuan, Wilson Wai Lun Fung et al. · 2009 · 1.7K citations
Silberschatz · 1991 · 1.6K citations
Dean M. Tullsen, Susan J. Eggers, Henry M. Levy · 1995 · 1.4K citations · Full text
Large-scale cluster management at Google with Borg
Abhishek Verma, Luis Pedrosa, Madhukar Korupolu et al. · 2015 · 1.3K citations · Full text