Cascaded DMA controller for speedup of indirect memory access in irregular applications

Tomoya Kashimata, Toshiaki Kitamura, Keiji Kimura, Hironori Kasahara

研究成果: Conference contribution

抄録

Indirect memory accesses caused by sparse linear algebra calculations are widely used in important real applications. However, they also cause serious inefficient memory accesses and pipeline stalls resulting low execution efficiency even with high memory bandwidth and much computational resource. One of the important issues of indirect memory accesses, such as accessing A[B[i]], is it requires two succeeding different memory accesses: the index loads (B[i]) and the following data element accesses (A[B[i]]). To overcome this situation, we propose the Cascaded-DMAC (CDMAC). This CDMAC is intended to be attached in each core of a multicore chip in addition to a CPU core, a vector accelerator, and a local data memory. It performs data transfers between an off-chip main memory and an in-core local data memory, which provides data to the accelerator. The key idea of the CDMAC is cascading two DMACs so that the first one loads indices, then the second one accesses data elements by using these indices. Thus, this organization realizes the autonomous indirect memory accesses by giving an index array and an element array, and obtains the efficient SIMD computations by lining up the sparse data into the local data memory. We implemented a multicore processor having the proposed CDMAC on an FPGA board. The evaluation result of sparse matrix-vector multiplications on the FPGA shows that the CDMAC achieves 17x speedup at most compared with the CPU data transfer.

本文言語English
ホスト出版物のタイトル2019 IEEE/ACM 9th Workshop on Irregular Applications
ホスト出版物のサブタイトルArchitectures and Algorithms, IA3 2019
出版社Institute of Electrical and Electronics Engineers Inc.
ページ71-76
ページ数6
ISBN(電子版)9781728159874
DOI
出版ステータスPublished - 2019 11
イベント9th IEEE/ACM Workshop on Irregular Applications: Architectures and Algorithms, IA3 2019 - Denver, United States
継続期間: 2019 11 18 → …

出版物シリーズ

名前2019 IEEE/ACM 9th Workshop on Irregular Applications: Architectures and Algorithms, IA3 2019

Conference

Conference9th IEEE/ACM Workshop on Irregular Applications: Architectures and Algorithms, IA3 2019
国/地域United States
CityDenver
Period19/11/18 → …

ASJC Scopus subject areas

  • ソフトウェア
  • 計算数学
  • ハードウェアとアーキテクチャ

フィンガープリント

「Cascaded DMA controller for speedup of indirect memory access in irregular applications」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル