Skip to Main content Skip to Navigation
Documents associated with scientific events

TC-CIM: Empowering Tensor Comprehensions for Computing-In-Memory

Abstract : Memristor-based, non-von-Neumann architectures performing tensor operations directly in memory are a promising approach to address the ever-increasing demand for energy-efficient, high-throughput hardware accelerators for Machine Learning (ML) inference. A major challenge for the programmability and exploitation of such Computing-In-Memory (CIM) architectures consists in the efficient mapping of tensor operations from high-level ML frameworks to fixed-function hardware blocks implementing in-memory computations. We demonstrate the programmability of memristor-based accelerators with TC-CIM, a fully-automatic, end-to-end compilation flow from Tensor Comprehensions, a mathematical notation for tensor operations, to fixed-function memristor-based hardware blocks. Operations suitable for acceleration are identified using Loop Tactics, a declarative framework to describe computational patterns in a poly-hedral representation. We evaluate our compilation flow on a system-level simulator based on Gem5, incorporating crossbar arrays of memristive devices. Our results show that TC-CIM reliably recognizes tensor operations commonly used in ML workloads across multiple benchmarks in order to offload these operations to the accelerator.
Document type :
Documents associated with scientific events
Complete list of metadatas

Cited literature [51 references]  Display  Hide  Download

https://hal.inria.fr/hal-02441163
Contributor : Andi Drebes <>
Submitted on : Wednesday, January 15, 2020 - 4:08:31 PM
Last modification on : Tuesday, September 22, 2020 - 3:57:39 AM
Long-term archiving on: : Thursday, April 16, 2020 - 6:06:26 PM

File

IMPACT_2020_paper_2.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02441163, version 1

Collections

Citation

Andi Drebes, Lorenzo Chelini, Oleksandr Zinenko, Albert Cohen, Henk Corporaal, et al.. TC-CIM: Empowering Tensor Comprehensions for Computing-In-Memory. IMPACT 2020 - 10th International Workshop on Polyhedral Compilation Techniques, Jan 2020, Bologna, Italy. ⟨hal-02441163⟩

Share

Metrics

Record views

198

Files downloads

416