A performance analysis framework for optimizing OpenCL applications on FPGAs
Authors:
- 1 March 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 114-125
Abstract
Recently, FPGA vendors such as Altera and Xilinx have released OpenCL SDK for programming FPGAs. However, the architecture of FPGA is significantly different from that of CPU/GPU, for which OpenCL is originally designed. Tuning the OpenCL code for good performance on FPGAs is still an open problem, since the existing OpenCL tools and models designed for CPUs/GPUs are not directly applicable to FPGAs. In the paper, we present an FPGA-based performance analysis framework that can shed light on the performance bottleneck and thus guide the code tuning for OpenCL applications on FPGAs. Particularly, we leverage static and dynamic analysis to develop an analytical performance model, which has captured the key architectural features of FPGA abstractions under OpenCL. Then, we provide four programmer-interpretable metrics to quantify the performance potentials of the OpenCL program with input optimization combination for the next optimization step. We evaluate our framework with a number of user cases, and demonstrate that 1) our analytical performance model can accurately predict the performance of OpenCL programs with different optimization combinations on FPGAs, and 2) our tool can be used to effectively guide the code tuning on alleviating the performance bottleneck.Keywords
This publication has 28 references indexed in Scilit:
- In-cache query co-processing on coupled CPU-GPU architecturesProceedings of the VLDB Endowment, 2014
- A reconfigurable fabric for accelerating large-scale datacenter services2014
- Performance Modeling for FPGAs: Extending the Roofline Model with High-Level Synthesis ToolsInternational Journal of Reconfigurable Computing, 2013
- Revisiting co-processing for hash joins on the coupled CPU-GPU architectureProceedings of the VLDB Endowment, 2013
- clSpMV2012
- Optimizing MapReduce for GPUs with effective shared memory usage2012
- A performance analysis framework for identifying potential benefits in GPGPU applications2012
- Performance of partial reconfiguration in FPGA systemsACM Transactions on Reconfigurable Technology and Systems, 2011
- Performance Analysis Framework for High-Level Language Applications in Reconfigurable ComputingACM Transactions on Reconfigurable Technology and Systems, 2010
- Accurate Area, Time and Power Models for FPGA-Based ImplementationsJournal of Signal Processing Systems, 2009