8000 GitHub - tanvisharma/WWWtoCiM: WWW: What, When, Where to Compute-in-Memory for Efficient Matrix Multiplication during Machine Learning Inference
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

WWW: What, When, Where to Compute-in-Memory for Efficient Matrix Multiplication during Machine Learning Inference

Notifications You must be signed in to change notification settings

tanvisharma/WWWtoCiM

Repository files navigation

WWW

Welcome to the repository for "What, When, Where to Compute-in-Memory for Efficient Matrix Multiplication during Machine Learning Inference" by Tanvi Sharma, Mustafa Ali, Indranil Chakroborty and Kaushik Roy. WWW utilizes already existing infrastructure from Timeloop/Accelergy to analytically evaluate different compute in memory designs when integrated in the memory hierarchy of a tensorcore like architecture.

image

Overview

WWW compares SRAM based Compute-in-Memory designs (referred as primitives in this work) by abstracting them in terms of the following template:

image

and deciding the dataflow using our priority based algorithm for each CiM primitive, when integrated at the shared memory and register file level.

Contents

This repository contains the setup for

  1. Timeloopfe infrastructure (accelergy-timeloop-infrastructure/ and timeloop-accelergy-execises/)
  2. Priority based algorithm (www-cim/constraints/)
  3. Scripts used to calculate final performance metrics and plot graphs (www-cim/post-process)

For the paper, please visit WWW.

About

WWW: What, When, Where to Compute-in-Memory for Efficient Matrix Multiplication during Machine Learning Inference

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
0