Abstract | ||
---|---|---|
As chip multiprocessors accommodate a growing number of cores, they demand interconnection networks that simultaneously provide low latency, high bandwidth, and low power. Our goal is to provide a comprehensive study of the interactions between the interconnection network and the memory hierarchy to enable a better co-design of both components. We explore the implications of the interconnect choice on overall performance by comparing the behaviour of three topologies (mesh, torus, and ring) and their concentrated versions. Simply choosing the concentrated mesh over the ring improves performance by over 40% in a 64-core chip. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1016/j.micpro.2016.01.005 | Microprocessors and Microsystems |
Keywords | Field | DocType |
Interconnection networks,Chip multiprocessor,Topology,Mesh,Torus,Ring | Memory hierarchy,Computer science,Parallel computing,Network on a chip,Chip,Network topology,Real-time computing,Latency (engineering),Interconnection,Memory controller,Cost efficiency,Distributed computing | Journal |
Volume | ISSN | Citations |
42 | 0141-9331 | 2 |
PageRank | References | Authors |
0.42 | 28 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Marta Ortín-Obón | 1 | 13 | 2.73 |
Darío Suárez-Gracia | 2 | 76 | 12.06 |
Maria Villarroya-Gaudo | 3 | 4 | 1.75 |
Cruz Izu | 4 | 149 | 23.41 |
Víctor Viñals Yúfera | 5 | 11 | 2.25 |