gtx 1070 very poor opencl benchmark score

#Gtx 1070 very poor opencl benchmark score code#

KTT was configured to perform exhaustive exploration of tuning spaces on each GPU under our test with profiling switched on. We ran them on four GPUs: GeForce GTX 680, GeForce GTX 750, GeForce GTX 1070 and GeForce RTX 2080. Raw autotuning data: Computation time and performance counters were measured for five benchmarks (GEMM, Convolution, Matrix transposition, 3D Coulomb summation and n-body) bundled with KTT.

Prediction models: Using our scripts, also available in KTT repository at Github, we trained models with the raw tuning data. They also differ in sizes of their tuning spaces. These benchmarks cover a wide range of computational problems: computing convolution, Coulomb summation in three dimensions, matrix multiplication, matrix transposition and N-body problem. KTT is freely available, five benchmarks are also there in ’examples’ folder. Raw autotuning data: Using our autotuning framework Kernel Tuning Toolkit (KTT), we measured computation time and collected hardware performance counters for whole tuning spaces of five benchmark CUDA codes on four GPUs. These data might benefit researchers who want to reproduce or build on our research.Īuto-tuning GPU kernels using hardware performance counters Here, we provide the models themselves and describe the scripts we implemented for their training. These scripts, freely available with our other codes, make it easier to experiment with search algorithms and compare them in a robust and reproducible way.ĭuring our research, we generated models for predicting values of performance counters from values of tuning parameters of our benchmarks. In particular, the script that simulates the tuning, i.e., replaces time-demanding compiling and executing the tuned kernels with a quick reading of the computation time from our measured data, makes it possible to inspect the convergence of tuning search over a large number of experiments. Moreover, we describe the scripts we used for robust evaluation of our searcher and comparison to others in detail.

#Gtx 1070 very poor opencl benchmark score code#

These data, which we provide here, might benefit research of search algorithms for the tuning spaces of GPU codes or research of relation between applied code optimization, hardware performance counters, and GPU kernels’ performance. With our framework Kernel Tuning Toolkit, freely available at Github, we measured computation times and hardware performance counters on several GPUs for the complete tuning spaces of five benchmarks. We have used them during the development and evaluation of a search method for tuning space proposed in. We have developed several autotuning benchmarks in CUDA that take into account performance-relevant source-code parameters and reach near peak-performance on various GPU architectures.