Compiling HPC Kernels for the REDEFINE CGRA
Authors:
- Kavitha T. MadhuKMKavitha T. Madhu
- Saptarsi DasSDSaptarsi Das
- Nalesh S.NSNalesh S.
- S. K. NandySNS. K. Nandy
- Ranjani NarayanRNRanjani Narayan
- 1 August 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 405-410
Abstract
In this paper, we present a compilation flow for HPC kernels on the REDEFINE coarse-grain reconfigurable architecture (CGRA). REDEFINE is a scalable macro-dataflow machine in which the compute elements (CEs) communicate through messages. REDEFINE offers the ability to exploit high degree of coarse-grain and pipeline parallelism. The CEs in REDEFINE are enhanced with reconfigurable macro data-paths called HyperCells that enable exploitation of fine-grain and pipeline parallelism at the level of basic instructions in static dataflow order. Application kernels that exhibit regularity in computations and memory accesses such as affine loop nests benefit from the architecture of HyperCell [1], [2]. The proposed compilation flow aims at exposing high degree of parallelism in loop nests in HPC application kernels using polyhedral analysis and generates meta-data to effectively utilize the computational resources in HyperCells. Memory is explicitly managed through compiler's assistance. We address the compilation challenges such as partitioning with load balancing, mapping and scheduling computations and management of operand data while targeting multiple HyperCells in the REDEFINE architecture. The proposed solution scales well meeting the performance objectives of HPC computing.Keywords
This publication has 9 references indexed in Scilit:
- A framework for post-silicon realization of arbitrary instruction extensions on reconfigurable data-pathsJournal of Systems Architecture, 2014
- Synthesis of Instruction Extensions on HyperCell, a reconfigurable datapath2014
- Automatic Generation of FPGA-Specific Pipelined AcceleratorsLecture Notes in Computer Science, 2011
- Instruction Set Innovations for the Convey HC-1 ComputerIEEE Micro, 2010
- REDEFINEACM Transactions on Embedded Computing Systems, 2009
- Partially reconfigurable interconnection network for dynamically reprogrammable resource array2009
- Generic routing rules and a scalable access enhancement for the Network-on-Chip RECONNECT2009
- A practical automatic polyhedral parallelizer and locality optimizer2008
- The MOLEN polymorphic processorIEEE Transactions on Computers, 2004