@article {lyons2012accelerator, title = {The accelerator store: A shared memory framework for accelerator-based systems}, journal = {ACM Transactions on Architecture. and Code Optimization}, volume = {8}, year = {2012}, pages = {1-22}, publisher = {Association for Computing Machinery (ACM)}, abstract = {

This paper presents the many-accelerator architecture, a design approach combining the scalability of homogeneous multi-core architectures and system-on-chip{\textquoteright}s high performance and power-efficient hardware accelerators. In preparation for systems containing tens or hundreds of accelerators, we characterize a diverse pool of accelerators and find each contains significant amounts of SRAM memory (up to 90\% of their area). We take advantage of this discovery and introduce the accelerator store, a scalable architectural component to minimize accelerator area by sharing its memories between accelerators. We evaluate the accelerator store for two applications and find significant system area reductions (30\%) in exchange for small overheads (2\% performance, 0\%--8\% energy). The paper also identifies new research directions enabled by the accelerator store and the many-accelerator architecture.

}, url = {https://doi.org/10.1145/2086696.2086727}, author = {Michael Lyons and Mark Hempstead and Gu Wei and David Brooks} }