低功率使用分段預充電技術的三位元內容可定址記憶體設計

Low Power TCAM Design using Segmented-Precharge Technique

范聖欣(Sheng-Hsin Fan)

指導教授 ： 賴飛羆

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2011年)

https://doi.org/10.6342/NTU.2011.00560

摘要

因為三位元內容可定址記憶體的快速搜尋特性，使得它在許多需要高速的設備中扮演著重要的特色，但是它的耗電量也非常的高。而在傳統的NOR-type三位元內容可定址記憶體的功率消耗主要是來自於動態功率消耗，在一次的搜尋當中，對每一條相符線都先做預先充電，之後再把比對錯誤的相符線做放電，由於比對錯誤的相符線遠比正確的占的多，因此大多數的相符線都被放電了，這造成了很大的功率消耗。在這篇論文中，我們根據IP字首存在三位元內容可定址記憶體裡的遮罩位元，當遮罩位元出現0時，有著連續性的特性，提出了一個分段式預充電技術應用在三位元內容可定址記憶體來減少動態功率的消耗。實驗結果顯示，在一般的情況下考慮功率和延遲，將相符線分成四段是最佳的分段，並且可以減少23.68%的功率消耗和額外的0.74ns延遲。

關鍵字

三位元內容可定址記憶體 ； 功率消耗 ； 相符線 ； 遮罩位元 ； 連續性 ； 分段式預充電 ； 延遲

並列摘要

Due to the characteristic of high-speed search in Ternary Content Addressable Memory (TCAM), it plays an important role in many high-speed devices, however it consumes too much power. The power consumption in conventional NOR-type TCAM is mainly from the dynamic power consumption, in which precharging all the match lines and discharging mismatched match lines during a search operation. Since the number of mismatched match lines is much more than the matched ones, most of the match lines are discharged leading to high power consumption. In this thesis, based on the characteristic of IP prefix stored in TCAM, the mask bit “0” has a continuity property when “0” appears. Therefore, we proposed a method called segmented-precharge technique in TCAM to reduce dynamic power consumption. The experimental results show that the proposed TCAM design with 4 segments is the most appropriate considering power and delay, which has 23.68% power saving and 0.74ns delay in average-case compared to the conventional TCAM.

並列關鍵字

Ternary Content Addressable Memory (TCAM) ； Power consumption ； Match line ； Mask bit ； Continuity ； Segmented-precharge ； Delay

參考文獻

[1] K. Pagiamtzis and A. Sheikholeslami, "Content-addressable memory (CAM) circuits and architectures: a tutorial and survey," Solid-State Circuits, IEEE Journal of , vol.41, no.3, pp. 712- 727, March 2006.

[3] V. Chaudhary, T.-H. Chen, F. Sheerin, and L.T. Clark, "Critical race-free low-power nand match line content addressable memory tagged cache memory," Computers & Digital Techniques, IET , vol.2, no.1, pp.40-44, January 2008.

[5] M. Kobayashi, T. Murase, and A. Kuriyama, "A longest prefix match search engine for multi-gigabit IP processing," Communications, 2000. ICC 2000. 2000 IEEE International Conference on , vol.3, no., pp.1360-1364 vol.3, 2000.

[7] D.S. Vijayasarathi, M. Nourani, M.J. Akhbarizadeh, and P.T. Balsara, "Ripple-precharge TCAM: a low-power solution for network search engines," Computer Design: VLSI in Computers and Processors, 2005. ICCD 2005. Proceedings. 2005 IEEE International Conference on , vol., no., pp. 243- 248, 2-5 Oct. 2005.

[8] M.J. Akhbarizadeh, M. Nourani, and C.D. Cantrell, "Prefix segregation scheme for a TCAM-based IP forwarding engine," Micro, IEEE , vol.25, no.4, pp. 48- 63, July-Aug. 2005.