English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

OpenCL for FPGA enables developers to design FPGAs using a programming modelsimilar for processors. Recent works have shown that code optimization at theOpenCL level is important to achieve high computational efficiency. However,existing works either focus primarily on optimizing single kernels or solelydepend on channels to design multi-kernel pipelines. In this paper, we proposea source-to-source compiler framework, MKPipe, for optimizing multi-kernelworkloads in OpenCL for FPGA. Besides channels, we propose new schemes toenable multi-kernel pipelines. Our optimizing compiler employs a systematicapproach to explore the tradeoffs of these optimizations methods. To enablemore efficient overlapping between kernel execution, we also propose a novelworkitem/workgroup-id remapping technique. Furthermore, we propose newalgorithms for throughput balancing and resource balancing to tune theoptimizations upon individual kernels in the multi-kernel workloads. Ourresults show that our compiler-optimized multi-kernels achieve up to 3.6x (1.4xon average) speedup over the baseline, in which the kernels have already beenoptimized individually.

MKPipe: A Compiler Framework for Optimizing Multi-Kernel Workloads in OpenCL for FPGA