Determining the best set of optimizations to apply to a kernel to be executed on the graphics processing unit (GPU) is a challenging problem. There are large sets of possible optimization configurations that can be applied, and many applications have multiple kernels. Each kernel may require a specific configuration to achieve the best performance, and moving an application to new hardware often requires a new optimization configuration for each kernel.In this work, we apply optimizations to GPU code using HMPP, a high-level directive-based language and source-tosource compiler that can generate CUDA / OpenCL code. However, programming with high-level languages may mean a loss of performance compared to using low-level languages. Our work shows that it is possible to improve the performance of a high-level language by using auto-tuning. We perform auto-tuning on a large optimization space on GPU kernels, focusing on loop permutation, loop unrolling, tiling, and specifying which loop(s) to parallelize, and show results on convolution kernels, codes in the PolyBench suite, and an implementation of belief propagation for stereo vision. The results show that our auto-tuned HMPP-generated implementations are significantly faster than the default HMPP implementation and can meet or exceed the performance of manually coded CUDA / OpenCL implementations.

Abstract

The QuantLib library is a popular library used for many areas of computational finance. In this work, the parallel processing power of the GPU is used to accelerate QuantLib financial applications. Black-Scholes, Monte-Carlo, Bonds, and Repo code paths in QuantLib are accelerated using hand-written CUDA and OpenCL codes specifically targeted for the GPU. Additionally, HMPP and OpenACC versions of the applications were created to drive the automatic generation of GPU code from sequential code. The results demonstrate a significant speedup for each code using each parallelization method. We were also able to increase the speedup of HMPP-generated code with auto-tuning.

Abstract. A CUDA kernel will utilize high-latency local memory for storage when there are not enough registers to hold the required data or if the data is an array that is accessed using a variable index within a loop. However, accesses from local memory take longer than accesses from registers and shared memory, so it is desirable to minimize the use of local memory. This paper contains an analysis of strategies used to reduce the use of local memory in a CUDA implementation of belief propagation for stereo processing. We perform experiments using registers as well as shared memory as alternate locations for data initially placed in local memory, and then develop a hybrid implementation that allows the programmer to store an adjustable amount of data in shared, register, and local memory. We show results of running our optimized implementations on two different stereo sets and across three generations of nVidia GPUs, and introduce an auto-tuning implementation that generates an optimized belief propagation implementation on any input stereo set on any CUDA-capable GPU.

No abstract