Title :
Decreasing the power consumption of content-addressable memory in the dataflow parallel computing system
Author :
Levchenko, N.N. ; Okunev, A.S. ; Yakhontov, D.E. ; Zmejev, D.N.
Author_Institution :
Inst. for Design Problems in Microelectron., Moscow, Russia
Abstract :
This article proposes the way to decrease the power consumption of content-addressable memory in dataflow parallel computing system. The proposed method is to split a memory word into several short segments and start the searching in the each successive segment only if there is match in the previous one. This approach allows decreasing the number of bit search operations by 77% according to software modeling results.
Keywords :
content-addressable storage; data flow computing; content-addressable memory; dataflow parallel computing system; decrease power consumption; memory word; software modeling results; CAM; DPCS; content-addressable memory; pipelining; power consumption;
Conference_Titel :
Design & Test Symposium, 2013 East-West
Conference_Location :
Rostov-on-Don
Print_ISBN :
978-1-4799-2095-2
DOI :
10.1109/EWDTS.2013.6673191