DocumentCode
259055
Title
Soft-error tolerant TCAMs for high-reliability packet classifications
Author
Syafalni, Infall ; Sasao, Tsutomu ; Xiaoqing Wen ; Holst, Stefan ; Miyase, Kohei
Author_Institution
Kyushu Inst. of Technol., Kitakyushu, Japan
fYear
2014
fDate
17-20 Nov. 2014
Firstpage
471
Lastpage
474
Abstract
In the internet, packets are classified by source and destination addresses, ports, and protocol type. Ternary content addressable memories (TCAMs) are often used to perform this operation. However, high-reliability packet classifiers in aerospace network applications require soft-error tolerant TCAMs to prevent system from errors caused by environmental factors such as high-energy radiations. This paper shows a soft-error tolerant TCAM (STTCAM), which enhances the reliability of TCAMs for soft-errors. The STTCAM randomly selects a search key to be evaluated. Then, parallel TCAMs detect single-bit flip errors. When the search key matches the last word, the STTCAM calculates the similarity of the search key to the TCAM word. If 99% of similarity is detected, then a suspected error is found and the STTCAM refreshes the TCAM words by using a backup ECC-SRAM. Experimental results show that the STTCAM improves TCAMs reliability significantly than the existing scheme. The STTCAM can be easily implemented and is useful for fault-tolerant packet classifiers.
Keywords
SRAM chips; content-addressable storage; fault tolerance; integrated circuit reliability; radiation hardening (electronics); ECC-SRAM; STTCAM; aerospace network application; environmental factor; fault-tolerant packet classifier; high-reliability packet classification; single-bit flip error detection; soft-error tolerant; ternary content addressable memory; Indexes; Internet; Ports (Computers); Probabilistic logic; Protocols; Random access memory; Reliability;
fLanguage
English
Publisher
ieee
Conference_Titel
Circuits and Systems (APCCAS), 2014 IEEE Asia Pacific Conference on
Conference_Location
Ishigaki
Type
conf
DOI
10.1109/APCCAS.2014.7032821
Filename
7032821
Link To Document