Single-Board SIMD Processors Using Gate-Array LSIs for Parallel Processing (Special Issue on ASICs for Automotive Electronics)
スポンサーリンク
概要
- 論文の詳細を見る
We have developed an SIMD processor on a double-height VME board. We achieved a good balance between cost and performance by combining four identical gatearray LSIs in the processor array with a 16-bit digital signal processor (DSP), standard dymanic random-access memories (DRAMs) and other peripherals. The gate-array LSIs have 168-bit processing elements (PEs), each containing a one-bit processing block and a serial multiplier. This PE structure offers high-level bit processing capability and peak performance of 512 million operations per second (MOPS) for 8-bit multiply and accumulate operations. Effective performance of more than 300 MOPS for 8-bit array data processing is achieved by using an LSI structure tuned to the DRAM access rate, although the processing speed is reduced by the DRAM access bottleneck. The LSIs also have two unique additional hardware structures that speed up various array data processes. One is an inter-PE routing register array for supporting a transmission, rotation and memory access path. The other is a tree-structure network for propagating operations among PEs. With these cost-effective structures, the SIMD processor is expected to be widely used for two-dimensional data processing, such as image processing and pattern recognition.
- 社団法人電子情報通信学会の論文
- 1993-12-25
著者
-
Sonehara Noboru
Ntt Human Interface Labs.
-
Sonehara Noboru
Ntt Human Interface Laboratories
-
Kondo Toshio
NTT LSI Laboratories
-
Kimura Yoshimasa
NTT Human Interface Laboratories
関連論文
- B-11-23 New "Filer" Tool for ATM Transmission Evaluation
- Single-Board SIMD Processors Using Gate-Array LSIs for Parallel Processing (Special Issue on ASICs for Automotive Electronics)
- Edge-based Binocular Stereopsis Algorithm : A Matching Mechanisim with Probabilistic Feedback
- Evaluation of Robustness in a Learning Algorithm that Minimizes Output Variation for Handprinted Kanji Pattern Recognition (Special Issue on Neurocomputing)