Sidebar: Scratchpad Based Communication Between CPUs and Accelerators
Ayoosh Bansal, Chance Coats, Evan Lissoos, Benjamin Schreiber

TL;DR
This paper introduces Sidebar, a shared buffer architecture enabling flexible neural network accelerators that coordinate with host processors, maintaining high performance and energy efficiency while allowing dynamic computation of evolving functions.
Contribution
The paper proposes Sidebar, a low-latency shared buffer that facilitates flexible accelerator design with host-accelerator cooperation, improving over DMA-dependent approaches.
Findings
Sidebar achieves performance comparable to fixed-function accelerators.
Sidebar maintains energy efficiency similar to fixed-function designs.
Flexible accelerators with Sidebar outperform DMA-based flexible designs.
Abstract
Hardware accelerators for neural networks have shown great promise for both performance and power. These accelerators are at their most efficient when optimized for a fixed functionality. But this inflexibility limits the longevity of the hardware itself as the underlying neural network algorithms and structures undergo improvements and changes. We propose and evaluate a flexible design paradigm for accelerators with a close coordination with host processors. The relatively static matrix operations are implemented in specialized accelerators while fast-evolving functions, such as activations, are computed on the host processor. This architecture is enabled by a low latency shared buffer we call Sidebar. Sidebar memory is shared between the accelerator and host, exists outside of program address space and holds intermediate data only. We show that a generalised DMA dependent flexible…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsParallel Computing and Optimization Techniques · Advanced Neural Network Applications · Advanced Memory and Neural Computing
