GOMA: Geometrically Optimal Mapping via Analytical Modeling for Spatial Accelerators

TL;DR

GOMA introduces a geometric abstraction and analytical modeling approach to efficiently find globally optimal GEMM mappings on spatial accelerators, significantly improving energy-delay performance and solution time.

Contribution

GOMA is the first framework to guarantee globally optimal GEMM mappings using an analytical energy model and geometric abstraction, enabling fast and optimal solutions.

Findings

GOMA achieves 2.24-4.24x energy-delay product improvement over state-of-the-art mappers.

GOMA reduces mapping solution time by 3.83-73.6x.

It provides exact analytical evaluation with O(1) complexity for GEMM mappings.

Abstract

General matrix multiplication (GEMM) on spatial accelerators is highly sensitive to mapping choices in both execution efficiency and energy consumption. However, the mapping space exhibits combinatorial explosion, which makes it extremely challenging to obtain optimal mappings within an acceptable time budget. Existing approaches typically face challenges: They often lack global-optimality guarantees and become prohibitively slow as the mapping space grows. To address these limitations, we propose \textsc{GOMA}, a geometric-abstraction-based, globally optimal GEMM mapping framework via analytical modeling, which achieves efficient solving while guaranteeing optimality. \textsc{GOMA} introduces, from first principles, a geometric abstraction for GEMM mapping, yielding an exact analytical energy objective with $O(1)$ evaluation for any given mapping. The objective is highly accurate. \textsc{GOMA} then formulates mapping selection as an integer optimization problem under hardware and mapping constraints, using the analytical energy model as the objective to automate mapping search. \textsc{GOMA} can quickly compute a global-optimal mapping for any (GEMM workload, target hardware) pair, achieving this for the first time in mapping space exploration. Experiments confirm that across representative accelerators and large language model prefill workloads, \textsc{GOMA} improves the energy--delay product (EDP) by $2.24$--$4.24\times$ over SOTA mappers, while accelerating time-to-solution by $3.83$--$73.6\times$.