DocumentCode
2997523
Title
FPGA Implementation of SRAM-based Ternary Content Addressable Memory
Author
Ullah, Zahid ; Jaiswal, Manish Kumar ; Chan, Y.C. ; Cheung, Ray C C
fYear
2012
fDate
21-25 May 2012
Firstpage
383
Lastpage
389
Abstract
Content Addressable Memory (CAM) is a special memory that accomplishes search operation in a single clock cycle but CAM has disadvantages like low bit density and high cost per bit. In this paper, we present an implementation of a 512 x 36 SRAM-based TCAM (SR-TCAM) on a Virtex-5 FPGA, which is the strength of SR-TCAM because currently classical TCAMs cannot be implemented on FPGA. We have used two synthesis optimizations (BRAM-AUTO and BRAM = BLOCK_POWER2) using BRAMs on FPGA. Thus, user can choose design parameters that are suitable for his application. The power data has been measured using Xilinx X-Power by using activities for search operations of SR-TCAM. We have computed power consumption by taking the average of 1000 search operations with 100 MHz clock speed to have better power estimation, which results, for one of the design, in an average power consumption of 2.11 mW. SR-TCAM exploits dense SRAM and achieves comparable search performance in two clock cycles. Thus, SR-TCAM is a feasible and practical alternative to traditional CAMs.
Keywords
SRAM chips; field programmable gate arrays; optimisation; BRAM-AUTO; FPGA implementation; SR-TCAM; SRAM-based TCAM; SRAM-based ternary content addressable memory; Virtex-5 FPGA; Xilinx X-Power; clock speed; content addressable memory; frequency 100 MHz; optimization synthesis; power 2.11 mW; power consumption; power estimation; search operations; Computer aided manufacturing; Computer architecture; Field programmable gate arrays; Generators; Microprocessors; Power demand; Random access memory; Content Addressable Memroy (CAM); FPGA; Latency; Matching Address (MA); Potential Matching Address (PMA); Power Consumption; SRAM; Ternary Content Addressable Memory (TCAM); Vertical Partitioning (VP);
fLanguage
English
Publisher
ieee
Conference_Titel
Parallel and Distributed Processing Symposium Workshops & PhD Forum (IPDPSW), 2012 IEEE 26th International
Conference_Location
Shanghai
Print_ISBN
978-1-4673-0974-5
Type
conf
DOI
10.1109/IPDPSW.2012.47
Filename
6270666
Link To Document