HeLEx: A Heterogeneous Layout Explorer for Spatial Elastic Coarse-Grained Reconfigurable Arrays

TL;DR

HeLEx is a framework that optimizes heterogeneous spatial elastic CGRAs by reducing operations, area, and power consumption, closely approaching the theoretical minimum while outperforming existing methods.

Contribution

HeLEx introduces a branch-and-bound based approach for designing optimized heterogeneous CGRAs tailored to specific data flow graphs.

Findings

Reduces number of operations by 68.7% on average.

Decreases CGRA area by nearly 70%.

Lowers power consumption by over 51%.

Abstract

We present HeLEx, a framework for determining the functional layout of heterogeneous spatially-configured elastic Coarse-Grained Reconfigurable Arrays (CGRAs). Given a collection of input data flow graphs (DFGs) and a target CGRA, the framework starts with a full layout in which every processing element (PE) supports every operation in the DFGs. It then employs a branch-and-bound (BB) search to eliminate operations out of PEs, ensuring that the input DFGs successfully map onto the resulting CGRAs, eventually returning an optimized heterogeneous CGRA. Experimental evaluation with 12 DFGs and 9 target CGRA sizes reveals that the framework reduces the number of operations by 68.7% on average, resulting in a reduction of CGRA area by almost 70% and of power by over 51%, all compared to the initial full layout. HeLEx generates CGRAs that are on average only within 6.2% of theoretically minimum CGRAs that support exactly the number of operations needed by the input DFGs. A comparison with functional layouts produced by two state-of-the-art frameworks indicates that HeLEx achieves better reduction in the number of operations, by up to 2.6X.