1. Programming Strategies for Irregular Algorithms on the Emu Chick
- Author
-
Jiajia Li, Ümit V. Çatalyürek, Rich Vuduc, Jeffrey Young, Bora Uçar, Eric R. Hein, Abdurrahman Yasar, Srinivas Eswar, Jason Riedy, Thomas M. Conte, Emu Technology, Georgia Institute of Technology [Atlanta], Pacific Northwest National Laboratory (PNNL), Optimisation des ressources : modèles, algorithmes et ordonnancement (ROMA), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de l'Informatique du Parallélisme (LIP), École normale supérieure - Lyon (ENS Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Lyon (ENS Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS), Laboratoire de l'Informatique du Parallélisme (LIP), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), This work partially was supported by NSF Grant ACI-1339745 (XScala), an IARPA contract, and the Defense Advanced Research Projects Agency (DARPA) under agreement #HR0011-13-2-0001. Any opinions, findings, conclusions, or recommendations in this paper are solely those of the authors and do not necessarily reflect the position or the policy of the sponsors. Thanks also to the Emu Technology team for support and debugging assistance with the Emu Chick prototype, and to the reviewers for their thoughtful comments., École normale supérieure de Lyon (ENS de Lyon)-Université Claude Bernard Lyon 1 (UCBL), and Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure de Lyon (ENS de Lyon)-Université Claude Bernard Lyon 1 (UCBL)
- Subjects
FOS: Computer and information sciences ,0301 basic medicine ,Computer science ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,Memory bandwidth ,02 engineering and technology ,Thread (computing) ,020202 computer hardware & architecture ,Computer Science Applications ,03 medical and health sciences ,030104 developmental biology ,Computational Theory and Mathematics ,Application-specific integrated circuit ,Computer Science - Distributed, Parallel, and Cluster Computing ,Hardware and Architecture ,Modeling and Simulation ,0202 electrical engineering, electronic engineering, information engineering ,Graph alignment ,Bandwidth (computing) ,Distributed, Parallel, and Cluster Computing (cs.DC) ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,Field-programmable gate array ,Algorithm ,Software ,Analysis of algorithms - Abstract
The Emu Chick prototype implements migratory memory-side processing in a novel hardware system. Rather than transferring large amounts of data across the system interconnect, the Emu Chick moves lightweight thread contexts to near-memory cores before the beginning of each remote memory read. Previous work has characterized the performance of the Chick prototype in terms of memory bandwidth and programming differences from more typical, non-migratory platforms, but there has not yet been an analysis of algorithms on this system. This work evaluates irregular algorithms that could benefit from the lightweight, memory-side processing of the Chick and demonstrates techniques and optimization strategies for achieving performance in sparse matrix-vector multiply operation (SpMV), breadth-first search (BFS), and graph alignment across up to eight distributed nodes encompassing 64 nodelets in the Chick system. We also define and justify relative metrics to compare prototype FPGA-based hardware with established ASIC architectures. The Chick currently supports up to 68x scaling for graph alignment, 80 MTEPS for BFS on balanced graphs, and 50% of measured STREAM bandwidth for SpMV.
- Published
- 2020