Evaluating Spatial Accelerator Architectures with Tiled Matrix-Matrix Multiplication

Technical Report · Wed Jul 14 04:00:00 EDT 2021

DOI:https://doi.org/10.2172/1808019· OSTI ID:1808019

Moon, Gordon E. ^[1]; Kwon, Hyoukjun ^[2]; Jeong, Geonhwa ^[2]; Chatarasi, Prasanth ^[2]; Rajamanickam, Sivasankaran ^[3]; Krishna, Tushar ^[2]

---

1. Korea Aerospace Univ., Goyang (Korea, Republic of)
2. Georgia Inst. of Technology, Atlanta, GA (United States)
3. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)

There is a growing interest in custom spatial accelerators for machine learning applications. These accelerators employ a spatial array of processing elements (PEs) interacting via custom buffer hierarchies and networks-on-chip. The efficiency of these accelerators comes from employing optimized dataflow (i.e., spatial/temporal partitioning of data across the PEs and fine-grained scheduling) strategies to optimize data reuse. The focus of this work is to evaluate these accelerator architectures using a tiled general matrix-matrix multiplication (GEMM) kernel. To do so, we develop a framework that finds optimized mappings (dataflow and tile sizes) for a tiled GEMM for a given spatial accelerator and workload combination, leveraging an analytical cost model for runtime and energy. Our evaluations over five spatial accelerators demonstrate that the tiled GEMM mappings systematically generated by our framework achieve high performance on various GEMM workloads and accelerators.

• Research Organization: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
• Sponsoring Organization: USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
• DOE Contract Number: AC04-94AL85000
• OSTI ID: 1808019
• Report Number(s): SAND2021-7479R; 697053
• Country of Publication: United States
• Language: English

Related Subjects

43 PARTICLE ACCELERATORS
97 MATHEMATICS AND COMPUTING
DNN accelerator
GEMM mapping
dataflow
spatial accelerator