Affiliation:
1. Technological Educational Institute of Crete, Crete, Greece
Abstract
Modern data centers increasingly employ FPGA-based heterogeneous acceleration platforms as a result of their great potential for continued performance and energy efficiency. Today, FPGAs provide more hardware parallelism than is possible with GPUs or CPUs, whereas C-like programming environments facilitate shorter development time, even close to software cycles. In this work, we address limitations and overheads in access and transfer of data to accelerators over common CPU-accelerator interconnects such as PCIe. We present three different FPGA accelerator dispatching methods for streaming applications (e.g., multimedia, vision computing). The first uses zero-copy data transfers and on-chip scratchpad memory (SPM) for energy efficiency, and the second uses also zero-copy but shared copy engines among different accelerator instances and local external memory. The third uses the processor’s memory management unit to acquire the physical address of user pages and uses scatter-gather data transfers with SPM. Even though all techniques exhibit advantages in terms of scalability and relieve the processor from control overheads through using integrated schedulers, the first method presents the best energy-efficient acceleration in streaming applications.
Publisher
Association for Computing Machinery (ACM)
Subject
Hardware and Architecture,Information Systems,Software
Reference33 articles.
1. Brad Brech Juan Rubio and Michael Hollinger. 2014. Data Engine for NoSQL —IBM Power Systems Edition. White Paper. IBM. https://www-304.ibm.com/webapp/set2/sas/f/capi/CAPI_FlashWhitePaper.pdf. Brad Brech Juan Rubio and Michael Hollinger. 2014. Data Engine for NoSQL —IBM Power Systems Edition. White Paper. IBM. https://www-304.ibm.com/webapp/set2/sas/f/capi/CAPI_FlashWhitePaper.pdf.
2. Instruction Set Innovations for the Convey HC-1 Computer
3. A quantitative analysis on microarchitectures of modern CPU-FPGA platforms
4. An Analysis of Accelerator Coupling in Heterogeneous Architectures
5. ffLink
Cited by
10 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献