Abstract
Coarse Grained Reconfigurable Arrays (CGRAs) or Architectures are a concept for hardware accelerators based on the idea of distributing workload over Processing Elements. These processors exploit instruction level parallelism, while being energy efficient due to their simplistic internal structure. However, the incorporation into a complete computing system raises severe challenges at the hardware and software level. This article evaluates a CGRA integrated into a control engineering environment targeting a Xilinx Zynq System on Chip (SoC) in detail. Besides the actual application execution performance, the practicability of the configuration toolchain is validated. Challenges of the real-world integration are discussed and practical insights are highlighted.
Highlights
In [16] we have presented a Coarse Grained Reconfigurable Arrays (CGRAs)-based hardware accelerator that was integrated into a Realtime (RT) target for control engineering purposes
As this paper focuses on the CGRA integration into the RT target, the core architecture is described only briefly
This section gives a detailed evaluation of the CGRA tool and performance
Summary
The CGRA is confronted with an integration into a real-world industrial environment. The resulting challenges for the accelerator that are not covered with usual (often synthetic) benchmark suites for accelerators. This contribution introduces the CGRA integration and gives a detailed analysis on how the system performs as well as the insights that can be drawn. It enhances previous research [16] especially by investigating the scalability of the tool and application runtime with the CGRA size and by proposing possible micro-architecture improvements allowing to implement larger CGRAs on the same FPGA.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have