What is it about?
The purpose of this paper is to exploit how can we easily reuse code previously written for execution on CPU and GPU, in FPGA. The objective is that a programmer without specific hardware design knowledge is capable of optimizing it, with a convenient exploitation of the design space.
Featured Image
Why is it important?
In the quest for high-performance computing, some contexts require accelerators without limitations (e.g., power), and for those cases GPUs or CPUs can be used, while others demand the use of low-power devices, thus opening new opportunities for FPGAs.
Perspectives
It is my opinion that FPGAs will grow in utilization and resources, namely in the new contexts of mobile computers used, for example, in autonomous driving or smartphones, that require low-power high-performance computing. In order to get there, we need tools like the one described in this paper to help programmers making the transition from typical programming languages to hardware design more quickly with acceptable levels of efficiency.
Gabriel Falcao
IT / University of Coimbra
Read the Original
This page is a summary of: Enhancing Design Space Exploration by Extending CPU/GPU Specifications onto FPGAs, ACM Transactions on Embedded Computing Systems, February 2015, ACM (Association for Computing Machinery),
DOI: 10.1145/2656207.
You can read the full text:
Contributors
The following have contributed to this page
