Abstract | ||
---|---|---|
Parallel mappings of Sparse Distributed Memory (SDM) are presented for a partial tree shape neurocomputer. The implemented partial tree shape neurocomputer called PARNEU is a scalable co-processor system to speed up soft computing applications. The used SDM algorithm benefits from the improved reading method and mask based activation. Two different mapping strategies are presented and analysed in detail The effect of SDM size is analysed and the system scalability is verified with speedup figures of different SDM networks. Almost linear speedup is achieved with reasonable size SDM algorithm. |
Year | Venue | Keywords |
---|---|---|
2000 | INTELLIGENT SYSTEMS | Sparse Distributed Memory, neurocomputer, parallel mapping |
Field | DocType | Citations |
Shared memory,Computer science,Parallel computing,Symmetric multiprocessor system,Distributed memory,Theoretical computer science,Multiprocessing,Distributed shared memory,Sparse distributed memory,Scalability | Conference | 0 |
PageRank | References | Authors |
0.34 | 1 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Pasi Kolinummi | 1 | 10 | 3.36 |
Kimmo Tikkanen | 2 | 0 | 0.34 |
Timo Hämäläinen | 3 | 1603 | 194.30 |
Jukka Saarinen | 4 | 264 | 46.21 |