(3.235.245.219) 您好!臺灣時間:2021/05/07 22:37
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:陳鍠元
研究生(外文):Hung-Yuan Chen
論文名稱:具成本考量漸進式繞徑晶片網路交換器設計
論文名稱(外文):Switch Design of Network on Chip with Cost-Efficient Incremental-Based Routing Algorithm
指導教授:許明華許明華引用關係
指導教授(外文):Ming-Hwa Sheu
學位類別:碩士
校院名稱:國立雲林科技大學
系所名稱:電子與資訊工程研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2009
畢業學年度:97
語文別:中文
論文頁數:115
中文關鍵詞:不規則繞徑演算法交換器設計晶片網路
外文關鍵詞:NOCIrregular routing algorithmSwitch design
相關次數:
  • 被引用被引用:0
  • 點閱點閱:92
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
近代積體電路設計中,市場產品的設計重點在縮減開發時間,因此矽智財(Silicon IP)的使用頻繁,隨著半導體製程越來越小,整合入系統單晶片(System on Chip)的IP數量日漸增多,使傳統的共享式匯流排(Share Bus)的頻寬已經不敷使用,因此晶片網路(Network on Chip) 成為開發的趨勢,由於IP的大小不相同,使晶片網路中將IP mapping後會形成不規則網路拓樸,所以發展一套具成本考量且有效因應各種不同拓樸的繞徑演算法成為當務之急。
本篇論文提出利用具成本考量漸進式繞徑演算法(CEIR , Cost-Efficent Incremental-Based Routing Algorithm)與晶片網路交換器設計。
CEIR是結合xy routing與source routing的繞徑方式,利用封包內虛擬的目的地座標,在虛擬目的點更新下一個目的點座標,繞徑會利用中繼點從source、intermediate、destination依序傳輸,因此CEIR沒有龐大的記憶體記錄routing path,更不需複雜的演算法進行不規則拓樸繞徑,此外arbiter的設計是採用分散式的公平仲裁,除了擁有傳統中央集權式公平性外,分散式架構在硬體設計上讓複雜度降低。
系統傳輸效率中,CEIR較OAPR效益提高13%,硬體cost方面,LUT(lookup table)最高可以降低99.23%的記錄資料量。
使用TSMC 0.13奈米製程下線一顆面積為1.658 x 1.658 mm2,消耗功率為36.66mW,經實際量測速度為133MHz的晶片網路交換器。
Nowaday, silicon intellectual properties (SIPs) are used to decrease time-to-market. With the advancement of semiconductor technology, many IPs can be integrated into a single chip lead to a large amount of data exchange. The bandwidth of traditional shared-bus interconnection cannot satisfy the requirement. Network on chip is a new paradigm to communicate IPs each other. After IP mapping, an irregular network topology is made because IP size is different. It is necessary to develop an efficient routing algorithm for irregular network topology.

In this thesis, we propose the cost-efficient incremental-based routing algorithm (CEIR) and implement an on-chip switch design.

The CEIR combines the advantages of x-y routing and source routing to reduce routing complexity and cost.Using the coordinates of pseudo destination in the packet,the next destination coordinates will put in pseudo destination switch.So CEIR doesn''t need huge memory to record routing path and the complicated algorithm to do irregular routing. The arbiter uses fair arbitration controller that has fair arbitration and low complex.

To compare with OAPR, the CEIR''s transmitting efficiency raises 13%.In the cost of hardware, the CEIR can save 99.23% recording data on LUT usage.

In TSMC 0.13 um process, the area is 1.658 x 1.658mm2 and the power consumption is 36.66mW.After measurement, the operating speed is 133MHz.
中文摘要 --------------------------------------------------- i
英文摘要 --------------------------------------------------- iii
誌謝 --------------------------------------------------- v
目錄 --------------------------------------------------- vi
表目錄 --------------------------------------------------- vii
圖目錄 --------------------------------------------------- viii
第一章 序論----------------------------------------------- 1
1.1 研究動機------------------------------------------- 1
1.2 研究目的------------------------------------------- 3
1.3 論文架構------------------------------------------- 3
第二章 晶片網路背景介紹及相關研究------------------------- 4
2.1 網路拓樸------------------------------------------- 6
2.2 基本繞徑方式--------------------------------------- 10
2.3 資料交換形式--------------------------------------- 13
2.4 封包格式 ------------------------------------------ 15
2.5 交換器種類與架構----------------------------------- 23
第三章 繞徑背景介紹及相關研究----------------------------- 28
3.1 Table Based的繞徑方式------------------------------ 28
3.2 靜態繞徑演算法+特殊限制---------------------------- 35
第四章 CEIR繞徑演算法介紹與模擬效能分析------------------- 51
4.1 CEIR繞徑演算法介紹--------------------------------- 51
4.2 CEIR封包格式與繞徑方式介紹------------------------- 55
4.3 CEIR模擬環境與驗證數據----------------------------- 62
第五章 硬體交換器之設計與成果----------------------------- 77
5.1 CEIR硬體交換器設計--------------------------------- 77
5.2 高效能交換器設計與晶片實現------------------------- 79
5.3 競賽成果與得獎作品--------------------------------- 91
第六章 結論----------------------------------------------- 100
6.1 總結----------------------------------------------- 100
6.2 未來方向------------------------------------------- 101
參考文獻 --------------------------------------------------- 102
[1]ITRS, Int’l Technology Roadmap for Semiconductors, http://public.itrs.net, 2009.

[2]M. Srinivasan, A. David, M. Paolo,C. Salvatore, L. Benini,G. De Micheli and L. Raffo,“Synthesis of Predictable Networks-on-Chip-Based Interconnect Architectures for Chip Multiprocessors” IEEE Trans. VLSI Systems, 15, (8), 2007.
[3] L. Benini and G. De Micheli, “Networks on chips: A new SoC paradigm,” IEEE Trans. Computers, 35, (1), 70–80, 2002.
[4]K.S. Ding, C.T. Ho, J.J. Tsay, “Matrix Transpose on Meshes with Wormhole and XY Routing,” Proc. of 6th Symposium on Parallel and Distributed Processing, 656–663, 1994.
[5] K.H. Chen and G.M. Chiu, “Fault-Tolerant Routing Algorithm for Meshes without Using Virtual Channels,” J. Information Science and Eng., 14, 765-783, 1998.
[6] J. Wu, “A Fault-Tolerant and Deadlock-Free Routing Protocol in 2D Meshes Based on Odd-Even Turn Model,” IEEE Trans. Computers, 52, (9), 1154-1169, 2003.
[7]S.Y. Lin, C.H. Huang, C.H. Chao,K.H. Huang, and A.Y. Wu, “Traffic-Balanced Routing Algorithm for Irregular Mesh-Based On-Chip Networks,” IEEE Trans. Computers, 57, (9), 2008.
[8]P. Zipf, H. Hinkelmann, A. Ashraf and M. Glesner, “A switch architecture and signal synchronization for GALS system-on-chips,” In Proc. of 17th Symposium on Integrated Circuits and Systems Design, 210-215, 2004.
[9]Cisco System, Inc., “EIGRP, enhanced interior gateway routing protocol,” http://www.cisco.com/univercd/cc/td/doc/cisintwk/ito_doc/en_igrp.htm
[10] Cisco System, Inc., “EIGRP, enhanced interior gateway routing protocol,” http://www.cisco.com/univercd/cc/td/doc/cisintwk/ito_doc/en_igrp.htm
[11] W. J. Dally and C. L. Seitz, “Deadlock-Free Message Routing in Multiprocessor Interconnection Networks,” IEEE Trans. Computers, C-36, (5), 547-553, 1987.
[12] L. M. Ni and P. K. McKinley, “A Survey of Wormhole Routing Techniques in Direct Networks,” IEEE Trans. Computers, 26, 62-76, 1993.
[13] TCP/IP, http://lips.lis.ntu.edu.tw/ytchiang/study/others/tcpip/basic.htm
[14] M. I. KAROL , M. G. HLUCHVJ and S. P. MORGAN, “Input Versus Output Queueing on a Space-Division Packet Switch.” Proc. IEEE global telecommunications conl., 19.4.1-19.4.5, 1986.
[15] M. Karol, M. Hluchyj, and S. Morgan, “Input versus output queueing on
a space-division switch,” IEEE Trans. Computers, 35, 1347–1356, 1987.
[16] M. Lin and N. Mckeown, “The Throughput of A Buffered Crossbar
Switch,” IEEE Communications Letters, 9, (5), 465–467, 2005.
[17] J. Flich, A. Mejia, P. Lopez, and J. Duato, “Region-based routing:An ef_cient routing mechanism to tackle unreliable hardware in network on chips,” Proc. Int. Symp. Network-on-Chip, 183-194, 2007.
[18] A. Mejia, J. Flich, J. Duato, S. Reinemo, and T. Skeie, “Segment-Based routing: An efficient fault-tolerant routing algorithm for meshes and tori,” 20th IPDPS, 2006.
[19] M.E. Gomez, J. Duato, J. Flich, P. Lopez, A. Robles, N.A. Nordbotten, T. Skeie, and O. Lysne, “A New Adaptive Fault-Tolerant Routing Methodology for Direct Networks,” HiPC, 462-473, 2004.
[20]G..M. Chiu, “The Odd-Even Turn Model for Adaptive Routing,” IEEE Trans. Parallel and Distrbuted Systems, 11, (7), 729-737, 2000.
[21] J. Flich and J. Duto, “Logic-Based Distributed Routing for NoCs,” IEEE Computer Architecture Letters. 13-16, 2007.
[22] S.W Yang, M.H. Sheu, C.K. Yeh, C.Y. Wen, C.C. Lin and W.K. Tsai, “Fast Fair Crossbar Scheduler for On-chip Router,” IEEE International Symposium on Circuits and Systems, 385-388, 2007.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔