A System Level Compiler for Massively-Parallel, Spatial, Dataflow Architectures
By: Dirk Van Essendelft , Patrick Wingo , Terry Jordan and more
Potential Business Impact:
Lets computers run faster on new chips.
We have developed a novel compiler called the Multiple-Architecture Compiler for Advanced Computing Hardware (MACH) designed specifically for massively-parallel, spatial, dataflow architectures like the Wafer Scale Engine. Additionally, MACH can execute code on traditional unified-memory devices. MACH addresses the complexities in compiling for spatial architectures through a conceptual Virtual Machine, a flexible domain-specific language, and a compiler that can lower high-level languages to machine-specific code in compliance with the Virtual Machine concept. While MACH is designed to be operable on several architectures and provide the flexibility for several standard and user-defined data mappings, we introduce the concept with dense tensor examples from NumPy and show lowering to the Wafer Scale Engine by targeting Cerebras' hardware specific languages.
Similar Papers
From Loop Nests to Silicon: Mapping AI Workloads onto AMD NPUs with MLIR-AIR
Computation and Language
Makes computers use special chips much faster.
SPADA: A Spatial Dataflow Architecture Programming Language
Distributed, Parallel, and Cluster Computing
Makes super-fast computers easier to program.
From PyTorch to Calyx: An Open-Source Compiler Toolchain for ML Accelerators
Hardware Architecture
Turns AI code into computer chips.