Union: A Unified HW-SW Co-Design Ecosystem in MLIR for Evaluating Tensor Operationson Spatial Accelerators

Conference · Mon Oct 18 04:00:00 EDT 2021

OSTI ID:1972822

Jeong, Geonhwa ^[1]; Kestor, Gokcen ^[2]; Chatarasi, Prasanth ^[3]; Parashar, Angshuman ^[4]; Tsa, Po-An ^[4]; Rajamanickam, Siva ^[5]; Gioiosa, Roberto ^[2]; Krishna, Tushar ^[1]

---

1. Georgia Institute of Technology
2. BATTELLE (PACIFIC NW LAB)
3. IBM
4. NVIDIA
5. sandia

To meet the extreme compute demands for deep learning across commercial and scientific applications, dataflow accelerators are becoming increasingly popular. While these“domain-specific” accelerators are not fully programmable like CPUs and GPUs, they retain varying levels of flexibility with respect to data orchestration, i.e., dataflow and tiling optimizations to enhance efficiency. There are several challenges when designing new algorithms and mapping approaches to execute the algorithms for a target problem on new hardware. Previous works have addressed these challenges individually. To address this challenge as a whole, in this work, we present an HW-SW co-design ecosystem for spatial accelerators called Union within the popular MLIR compiler infrastructure. Our framework allows exploring different algorithms and their mappings on several accelerator cost models. Union also includes a plug-and-play library of accelerator cost models and mappers which can easily be extended. The algorithms and accelerator cost models are connected via a novel mapping abstraction that captures the map space of spatial accelerators which can be systematically pruned based on constraints from the hardware, workload, and mapper. We demonstrate the value of Union for the community with several case studies which examine offloading different tensor operations (CONV/GEMM/Tensor Contraction) on diverse accelerator architectures using different mapping schemes.

• Research Organization: Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
• Sponsoring Organization: USDOE
• DOE Contract Number: AC05-76RL01830
• OSTI ID: 1972822
• Report Number(s): PNNL-SA-168329
• Country of Publication: United States
• Language: English