论文信息 - Algorithm and Data Optimization Techniques for Scaling to Massively Threaded Systems

Algorithm and Data Optimization Techniques for Scaling to Massively Threaded Systems

A study of the implementation patterns among massively threaded applications for many-core GPUs reveals that each of the seven most commonly used algorithm and data optimization techniques can enhance the performance of applicable kernels by 2 to 10× in current processors while also improving future scalability.

[1] Ulf Assarsson,et al. Efficient stream compaction on wide SIMD many-core architectures , 2009, High Performance Graphics.

[2] Klaus Schulten,et al. Fast molecular electrostatics algorithms on GPUs , 2011 .

[3] Liwen Chang,et al. Optimization and architecture effects on GPU computing workload performance , 2012, 2012 Innovative Parallel Computing (InPar).

[4] Wen-mei W. Hwu,et al. DL: A data layout transformation system for heterogeneous computing , 2012, 2012 Innovative Parallel Computing (InPar).

[5] Wen-mei W. Hwu,et al. GPU Computing Gems Jade Edition , 2011 .