1. [Publications](/publications)
2. Sparseloop: An Analytical, Energy-Focused Design Space Exploration Methodology for Sparse Tensor Accelerators
 
 # Sparseloop: An Analytical, Energy-Focused Design Space Exploration Methodology for Sparse Tensor Accelerators

  ![Publication image](/sites/default/files/styles/wide/public/default_images/default.jpeg?itok=qUFsuJCP "Publication image")

 This paper presents Sparseloop, the first infrastructure that implements an analytical design space exploration methodology for sparse tensor accelerators. Sparseloop comprehends a wide set of architecture specifications including various sparse optimization features such as compressed tensor storage. Using these specifications, Sparseloop can calculate a design's energy efficiency while accounting for both optimization savings and metadata overhead at each storage and compute level of the architecture using stochastic tensor density models. We validate Sparseloop on a well-known accelerator design and achieve ~99% accuracy in terms of runtime activities (e.g., compressed memory accesses). We also present a case study that highlights the key factors (e.g., uncompressed traffic, data density) that affect sparse optimization features' impact on energy efficiency. Tool available at: <https://github.com/NVlabs/timeloop>.



 ## Authors



Yannan Nellie Wu (MIT)

[Po-An Tsai](/person/po-an-tsai)

[Angshuman Parashar](/person/angshuman-parashar)

Vivienne Sze (MIT)

[Joel Emer](/person/joel-emer)

 

 

 ## Publication Date



Wednesday, April 28, 2021

 

 ## Published in



[International Symposium on Performance Analysis of Systems and Software (ISPASS)](https://ieeexplore.ieee.org/document/9408213)

 

 ## Research Area



[Computer Architecture](/research-area/computer-architecture)

 

 

 ## External Links



[IEEE Digital Library](https://ieeexplore.ieee.org/document/9408213)

 

 

 ## Uploaded Files



[Published manuscript](https://d1qx31qr3h6wln.cloudfront.net/publications/ISPASS_2021_Sparseloop.pdf "Open file in new window")473.34 KB

 

 

 ## Copyright



This material is posted here with permission of the IEEE. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to <pubs-permissions@ieee.org>.