Title :
An Intelligent Multi-Port Memory
Author :
Wang Zuo ; Zuo, Wang ; Li Jiaxing
Author_Institution :
Sch. of Comput. Sci. & Technol., Beijing Inst. of Technol., Beijing
Abstract :
Multi-processors systems provide excellent computing ability for varied AI applications. Because processors may issue simultaneous accesses to the shared memory system, multi-port memory are widely used current multi-processor systems. This paper proposes an intelligent multi-port memory. With the help of port-priority and r/w-priority, our multi-port memory can resolve both write-write conflict and read-write conflict. When write-write conflict occurs, the port with the highest priority can execute its write operation; when read-write conflict occurs, the multi-port memory intelligently chooses read-then-write or write-then-read according to r/w-priority. Besides, direct data-switch paths between ports are implemented by utilizing the read-write conflict. We verify our design on Xilinx Virtex-E FPGA. Experiment results show that our design works well. Experiment results also show that, as the port number N increases, the cell cost increases significantly. This restricts the reasonable port number to rather small values in practice.
Keywords :
field programmable gate arrays; shared memory systems; Xilinx Virtex-E FPGA; intelligent multiport memory; multiprocessors systems; read-then-write; shared memory system; write-then-read; Application software; Artificial intelligence; Communication system control; Computer science; Costs; Field programmable gate arrays; Information technology; Multiprocessing systems; Read-write memory; Signal resolution; conflict; multi-port memory; port number; priority;
Conference_Titel :
Intelligent Information Technology Application Workshops, 2008. IITAW '08. International Symposium on
Conference_Location :
Shanghai
Print_ISBN :
978-0-7695-3505-0
DOI :
10.1109/IITA.Workshops.2008.231