Name
Abstract | ||
|---|---|---|
Interconnection network (ICN) has been playing an important role in parallel systems, since it greatly affects the system-level organization as well as its inherent communication capability. State-of-the-art parallel systems employ a huge number of computing nodes that are connected by an interconnection network. In general, an interconnection network shows nonlinear phenomena in its communication performance, most of them are caused by congestion. Thus, designing a large-scale parallel system requires sufficient discussions through repetitive simulation runs. This causes another problem in simulating a large-scale system within a reasonable cost. We have introduced the cellular automata principle in the ICN simulation and proposed an effective simulation model as a promising solution. However, the proposed method is insufficient in packet length and memory usage. This paper extends our prior simulation model to solve the problems. We firstly extend the existing cellular automata based model to match multiple-length packets. Furthermore, we discuss reduction methods of memory resources and propose a compressed buffer model and composite packet information. The proposed models have a good affinity to GPGPU technology and the GPU-based simulator accelerates simulation upto about 364 times from sequential execution on a single CPU. Furthermore, since the proposed models are applicable in the shared memory model, multithread implementation of the proposed methods achieve about 78 times speed-ups at the maximum. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1109/CANDAR.2017.52 | 2017 Fifth International Symposium on Computing and Networking (CANDAR) |
Keywords | Field | DocType |
interconnection networks,simulation,GPGPU,multithread | Cellular automaton,Computer science,Network packet,Automaton,Parallel computing,Field-programmable gate array,Network simulation,Acceleration,General-purpose computing on graphics processing units,Interconnection | Conference |
ISSN | ISBN | Citations |
2379-1888 | 978-1-5386-2088-5 | 0 |
PageRank | References | Authors |
0.34 | 4 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Takashi Yokota | 1 | 41 | 21.70 |
| Kanemitsu Ootsu | 2 | 44 | 23.90 |
| Takeshi Ohkawa | 3 | 21 | 16.24 |