LISA: Graph Neural Network based Portable Mapping on Spatial Accelerators

Spatial accelerators, such as Coarse-Grained Reconfigurable Arrays (CGRA), provide a promising pathway to scale the performance and power efficiency of computing systems. These accelerators depend on effective compilers to take advantage of the parallelism offered by the underlying architecture. Currently, the compilers are handcrafted for spatial accelerators, which is challenging from time to market perspective, especially with the rapid increase of diverse accelerators. In this paper, we present a portable compilation framework, called LISA, that can be tuned automatically to generate quality mapping for varied spatial accelerators. Our key contribution is to automatically identify the impact of the dataflow graph (DFG) structure characteristics (representing an application) on the mapping for a new accelerator. Towards this end, we abstract the DFG structure in graph attributes, use Graph Neural Network (GNN) to analyze the graph attributes, and identify the mapping impact for an accelerator architecture with an all-encompassing global view. Finally, we augment a simulated annealing-based mapping approach to take into account the impact of DFG structure in guiding the placement of the dataflow graph nodes and the routing of the dependencies on the accelerator. Our experimental evaluation concretely demonstrates the substantial benefit of our approach compared to the state-of-the-art solutions.

Automated summary

This paper presents a portable compilation framework called LISA, which automatically adjusts to generate high-quality mappings for various spatial accelerators. By using graph neural networks to analyze graph attributes and considering the impact of dataflow graph structure on node placement and dependency routing, an optimized mapping strategy is achieved.