Communication Lower Bound in Convolution Accelerators

November 08, 2019 Β· Declared Dead Β· πŸ› International Symposium on High-Performance Computer Architecture

πŸ‘» CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Xiaoming Chen, Yinhe Han, Yu Wang arXiv ID 1911.05662 Category cs.DC: Distributed Computing Cross-listed cs.AR Citations 35 Venue International Symposium on High-Performance Computer Architecture Last Checked 4 months ago
Abstract
In current convolutional neural network (CNN) accelerators, communication (i.e., memory access) dominates the energy consumption. This work provides comprehensive analysis and methodologies to minimize the communication for CNN accelerators. For the off-chip communication, we derive the theoretical lower bound for any convolutional layer and propose a dataflow to reach the lower bound. This fundamental problem has never been solved by prior studies. The on-chip communication is minimized based on an elaborate workload and storage mapping scheme. We in addition design a communication-optimal CNN accelerator architecture. Evaluations based on the 65nm technology demonstrate that the proposed architecture nearly reaches the theoretical minimum communication in a three-level memory hierarchy and it is computation dominant. The gap between the energy efficiency of our accelerator and the theoretical best value is only 37-87%.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Distributed Computing

Died the same way β€” πŸ‘» Ghosted