Barrier-Free Large-Scale Sparse Tensor Accelerator (BARISTA) For Convolutional Neural Networks Article Swipe

MAKE IMPACT

Ashish Gondimalla , Sree Charan Gundabolu , T. N. Vijaykumar , Mithuna Thottethodi · YOU? · · 2021 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2104.08734

Convolutional neural networks (CNNs) are emerging as powerful tools for visual recognition. Recent architecture proposals for sparse CNNs exploit zeros in the feature maps and filters for performance and energy without losing accuracy. Sparse architectures that exploit two-sided sparsity in both feature maps and filters have been studied only at small scales (e.g., 1K multiply-accumulate(MAC) units). However, to realize their advantages in full, the sparse architectures have to be scaled up to levels of the dense architectures (e.g., 32K MACs in the TPU). Such scaling is challenging since achieving reuse through broadcasts incurs implicit barrier cost raises the inter-related issues of load imbalance, buffering, and on-chip bandwidth demand. SparTen, a previous scheme, addresses one aspect of load balancing but not other aspects, nor the other issues of buffering and bandwidth. To that end, we propose the barrier-free large-scale sparse tensor accelerator (BARISTA). BARISTA (1) is the first architecture for scaling up sparse CNN accelerators; (2) reduces on-chip bandwidth demand by telescoping request-combining the input map requests and snarfing the filter requests; (3) reduces buffering via basic buffer sharing and avoids the ensuing barriers between consecutive input maps by coloring the output buffers; (4) load balances intra-filter work via dynamic round-robin work assignment; and (5) employs hierarchical buffering which achieves high cache bandwidth via a few, wide, shared buffers and low buffering via narrower, private buffers at the compute. Our simulations show that, on average, barista performs 5.4x, 2.2x, 1.7x, 2.5x better than a dense, a one-sided, a naively-scaled two-sided, and an iso-area two-sided architecture, respectively. Using 45-nm technology, ASIC synthesis of our RTL design for four clusters of 8K MACs at 1 GHz clock speed, reports 213 mm$^2$ area and 170 W power.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2104.08734

PDF

https://arxiv.org/pdf/2104.08734

OA Status

green

References

43

Related Works

20

OpenAlex ID

https://openalex.org/W3156219765

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W3156219765

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2104.08734

Digital Object Identifier

Title

Barrier-Free Large-Scale Sparse Tensor Accelerator (BARISTA) For Convolutional Neural Networks

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2021

Year of publication

Publication date

2021-04-18

Full publication date if available

Authors

Ashish Gondimalla, Sree Charan Gundabolu, T. N. Vijaykumar, Mithuna Thottethodi

List of authors in order

Landing page

https://arxiv.org/abs/2104.08734

Publisher landing page

PDF URL

https://arxiv.org/pdf/2104.08734

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2104.08734

Direct OA link when available

Concepts

Computer science, Exploit, Convolutional neural network, Bandwidth (computing), Parallel computing, Cache, Scaling, Reuse, Computer engineering, Memory bandwidth, Chip, Artificial intelligence, Computer network, Computer security, Mathematics, Ecology, Biology, Geometry, Telecommunications

Top concepts (fields/topics) attached by OpenAlex

Cited by

0

Total citation count in OpenAlex

References (count)

43

Number of works referenced by this work

Related works (count)

20

Other works algorithmically related by OpenAlex

Full payload