Abstract
The proliferation of processing hardware alternatives allows developers to use various customized computing platforms to run their applications in an optimal way. However, porting application code on custom hardware requires a lot of development and porting effort. This paper describes a heterogeneous computational platform (the ALMARVI execution platform) comprising of multiple communicating processors that allow easy programmability through an interface to OpenCL. The ALMARVI platform uses processing elements based on both VLIW and Transport Triggered Architectures (ρ-VEX and TCE cores, respectively). It can be implemented on Zynq devices such as the ZedBoard, and supports OpenCL by means of the pocl (Portable OpenCL) project and our ALMAIF interface specification. This allows developers to execute kernels transparently on either processing elements, thereby allowing to optimize execution time with minimal design and development effort.
Highlights
Image and video processing is an important enabler for a large number of application domains ranging from medical imaging to entertainment
The first supported custom processing units that supported by the platform are based on the Transport Triggered Architectures (TTA) paradigm [4]
They are generated by the TTA-based Co-design Environment (TCE), an open-source project maintained by Tampere University of Technology (TUT) [5]
Summary
Image and video processing is an important enabler for a large number of application domains ranging from medical imaging to entertainment. We present the ALMARVI execution platform which enables both hardware design flexibility as well as portable software deployment. The software stack allows for portable deployment of applications on each core based on an OpenCL programming layer [3] that has broad support in the industry for a whole range of computing platforms: multicore CPUs and GPUs, in addition to the upcoming standard of automatic hardware synthesis right from OpenCL code.
Published Version (Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have