BiCuts: A fast packet classification algorithm using bit-level cutting

Slides:



Advertisements
Similar presentations
Chapter Programming in C
Advertisements

FCU, Department of ECE, IC Design Research Lab. TEL: # 4945 Pre-SIm , Post-Sim.
McGraw-Hill/Irwin © 2003 The McGraw-Hill Companies, Inc.,All Rights Reserved. 肆 資料分析與表達.
Chapter 10 馬可夫鏈 緒言 如果讀者仔細觀察日常生活中所發生的 諸多事件,必然會發現有些事件的未來 發展或演變與該事件現階段的狀況全然 無關,這種事件稱為獨立試行過程 (process of independent trials) ;而另一些 事件則會受到該事件現階段的狀況影響。
建立使用案例敘述 --Use Case Narrative
: A-Sequence 星級 : ★★☆☆☆ 題組: Online-judge.uva.es PROBLEM SET Volume CIX 題號: Problem D : A-Sequence 解題者:薛祖淵 解題日期: 2006 年 2 月 21 日 題意:一開始先輸入一個.
:Word Morphing ★★☆☆☆ 題組: Problem Set Archive with Online Judge 題號: 10508:word morphing 解題者:楊家豪 解題日期: 2006 年 5 月 21 日 題意: 第一行給你兩個正整數, 第一個代表下面會出現幾個字串,
Teacher : Ing-Jer Huang TA : Chien-Hung Chen 2015/6/3 Course Embedded Systems : Principles and Implementations Weekly Preview Question CH3.5 ~ CH /10/31.
小綠人行人用交通號誌燈 指導教授:黃朝章 學生:951408陳宜歆 徐祐祥.
亂數產生器安全性評估 之統計測試 SEC HW7 姓名:翁玉芬 學號:
1.1 電腦的特性 電腦能夠快速處理資料:電腦可在一秒內處理數百萬個 基本運算,這是人腦所不能做到的。原本人腦一天的工 作量,交給電腦可能僅需幾分鐘的時間就處理完畢。 電腦能夠快速處理資料:電腦可在一秒內處理數百萬個 基本運算,這是人腦所不能做到的。原本人腦一天的工 作量,交給電腦可能僅需幾分鐘的時間就處理完畢。
STAT0_sampling Random Sampling  母體: Finite population & Infinity population  由一大小為 N 的有限母體中抽出一樣本數為 n 的樣 本,若每一樣本被抽出的機率是一樣的,這樣本稱 為隨機樣本 (random sample)
JAVA 程式設計與資料結構 第十四章 Linked List. Introduction Linked List 的結構就是將物件排成一列, 有點像是 Array ,但是我們卻無法直接經 由 index 得到其中的物件 在 Linked List 中,每一個點我們稱之為 node ,第一個 node.
各種線上電子資源的特異功能 STICnet 的 SDI 專題訂閱服務 2003/4/28 修改. 無論校內外皆可使用。連線至
Monte Carlo Simulation Part.2 Metropolis Algorithm Dept. Phys. Tunghai Univ. Numerical Methods C. T. Shih.
1 第四章 多變數函數的微分學 § 4.1 偏導數定義 定義 極限值 ■. 2 定理 極限值的基本定理 (1) 極限值的唯一性 : 若 存在,則 其值必為唯一。 (2) 若 且 ( 與 為常數 ) , 則 且 為常數且.
Wireless Protocol Bluetooth
Chapter 13 塑模靜態觀點:物件圖 Static View : Object Diagram.
Introduction to Java Programming Lecture 17 Abstract Classes & Interfaces.
: The largest Clique ★★★★☆ 題組: Contest Archive with Online Judge 題號: 11324: The largest Clique 解題者:李重儀 解題日期: 2008 年 11 月 24 日 題意: 簡單來說,給你一個 directed.
3-3 使用幾何繪圖工具 Flash 的幾何繪圖工具包括線段工具 (Line Tool) 、橢圓形工具 (Oval Tool) 、多邊星形 工具 (Rectangle Tool) 3 種。這些工具畫出 來的幾何圖形包括了筆畫線條和填色區域, 將它們適當地組合加上有技巧地變形與配 色, 不但比鉛筆工具簡單,
© The McGraw-Hill Companies, Inc., 2006© The McGraw-Hill Companies, Inc., 2007 Chapter 7 位址解析通訊協定及 反向位址解析通訊協定.
: Happy Number ★ ? 題組: Problem Set Archive with Online Judge 題號: 10591: Happy Number 解題者:陳瀅文 解題日期: 2006 年 6 月 6 日 題意:判斷一個正整數 N 是否為 Happy Number.
: Multisets and Sequences ★★★★☆ 題組: Problem Set Archive with Online Judge 題號: 11023: Multisets and Sequences 解題者:葉貫中 解題日期: 2007 年 4 月 24 日 題意:在這個題目中,我們要定義.
: Placing Lampposts ★★★☆☆ 題組: Problem Set Archive with Online Judge 題號: 10859: Placing Lampposts 解題者:陳志瑜 解題日期: 2011 年 5 月 10 日 題意:美化為 Dhaka City.
:Nuts for nuts..Nuts for nuts.. ★★★★☆ 題組: Problem Set Archive with Online Judge 題號: 10944:Nuts for nuts.. 解題者:楊家豪 解題日期: 2006 年 2 月 題意: 給定兩個正整數 x,y.
資料結構實習-一 參數傳遞.
Lecture 7 Sorting in Linear Time. Sorting in Linear Time2 7.1 Lower bounds for sorting 本節探討排序所耗用的時間複雜度下限。 任何一個以比較為基礎排序的演算法,排序 n 個元 素時至少耗用 Ω(nlogn) 次比較。
Dynamic Multi-signatures for Secure Autonomous Agents Panayiotis Kotzanikolaou Mike Burmester.
1 Introduction to Java Programming Lecture 2: Basics of Java Programming Spring 2008.
: A-Sequence ★★★☆☆ 題組: Problem Set Archive with Online Judge 題號: 10930: A-Sequence 解題者:陳盈村 解題日期: 2008 年 5 月 30 日 題意: A-Sequence 需符合以下的條件, 1 ≤ a.
1 Web of Science 利用指引 逢甲大學圖書館 參考服務組 單元六 個人化服務.
Section 4.2 Probability Models 機率模式. 由實驗看機率 實驗前先列出所有可能的實驗結果。 – 擲銅板:正面或反面。 – 擲骰子: 1~6 點。 – 擲骰子兩顆: (1,1),(1,2),(1,3),… 等 36 種。 決定每一個可能的實驗結果發生機率。 – 實驗後所有的實驗結果整理得到。
Teacher : Ing-Jer Huang TA : Chien-Hung Chen 2015/6/25 Course Embedded Systems : Principles and Implementations Weekly Preview Question CH 2.4~CH 2.6 &
函式 Function Part.2 東海大學物理系‧資訊教育 施奇廷. 遞迴( Recursion ) 函式可以「呼叫自己」,這種動作稱為 「遞迴」 此程式的執行結果相當於陷入無窮迴圈, 無法停止(只能按 Ctrl-C ) 這給我們一個暗示:函式的遞迴呼叫可以 達到部分迴圈的效果.
JAVA 程式設計與資料結構 第二十章 Searching. Sequential Searching Sequential Searching 是最簡單的一種搜尋法,此演 算法可應用在 Array 或是 Linked List 此等資料結構。 Sequential Searching 的 worst-case.
資料結構實習-二.
演算法 8-1 最大數及最小數找法 8-2 排序 8-3 二元搜尋法.
845: Gas Station Numbers ★★★ 題組: Problem Set Archive with Online Judge 題號: 845: Gas Station Numbers. 解題者:張維珊 解題日期: 2006 年 2 月 題意: 將輸入的數字,經過重新排列組合或旋轉數字,得到比原先的數字大,
Linguistics phonetic symbols. 先下載 IPA 字型檔案,執行安裝。 由於這個程式的字型目錄設定錯誤, 所以等重新開機時就會發現字型消失。 所以必須根據以下步驟來讓 Windows 加入 IPA 字型。
Structural Equation Modeling Chapter 6 CFA 根據每個因素有多重指標,以減少 測量誤差並可建立問卷的構念效度 驗證性因素分析.
網路連結 Internetworking.
Chapter 10 m-way 搜尋樹與B-Tree
: Function Overloading ★★★☆☆ 題組: Problem Set Archive with Online Judge 題號: 11032:Function Overloading 解題者:許智祺 解題日期: 2007 年 5 月 8 日 題意:判對輸入之數字是否為.
JAVA 程式設計與資料結構 第十六章 Hash Tables. Introduction Hash Tables 結構為一個 Array ,稱之為 Bucket array 。 如果想要新增一個物件,要根據這個物件的特性 將其加入 Hash Table 內。 Bucket Array 用 A 來代替,其.
網路介紹及其運用 講師陳炯勳. 5-2 IP 協定 ( 一 ) IP 協定運作 (1) – 網路成員:主機 (Host) 與路由器 (Router) – 路由表 – 電報傳輸運作.
概念性產品企劃書 呂學儒 李政翰.
無線通訊網路 Mac 層 TDM 通訊模式的操作與效能研究 專題生 : 林書弘、蔡逸祥、毛建翔、王政 華 指導教授 : 黃依賢.
1 Introduction to Java Programming Lecture 2: Basics of Java Programming Spring 2009.
專題成果報告 胺基酸功能預測開發環境 指導教授:歐昱言 邱彥豪 邱顯鈞.
5 重複迴圈 5.1 增減運算符號 增量運算符號 減量運算符號
: Problem E Antimatter Ray Clearcutting ★★★★☆ 題組: Problem Set Archive with Online Judge 題號: 11008: Problem E Antimatter Ray Clearcutting 解題者:林王智瑞.
第 6 章 迴圈結構 6-1 計數迴圈 6-1 計數迴圈 6-2 條件迴圈 6-2 條件迴圈 6-3 巢狀迴圈 6-3 巢狀迴圈 6-4 While/End While 迴圈 6-4 While/End While 迴圈 6-5 跳出迴圈 6-5 跳出迴圈 6-6 VB.NET 的錯誤處理 6-6 VB.NET.
Teacher : Ing-Jer Huang TA : Chien-Hung Chen 2015/6/30 Course Embedded Systems : Principles and Implementations Weekly Preview Question CH7.1~CH /12/26.
: Wine trading in Gergovia ★★☆☆☆ 題組: Contest Volumes with Online Judge 題號: 11054: Wine trading in Gergovia 解題者:劉洙愷 解題日期: 2008 年 2 月 29 日 題意:在 Gergovia.
: SAM I AM ★★★★☆ 題組: Contest Archive with Online Judge 題號: 11419: SAM I AM 解題者:李重儀 解題日期: 2008 年 9 月 11 日 題意: 簡單的說,就是一個長方形的廟裡面有敵人,然 後可以橫的方向開砲或縱向開砲,每次開砲可以.
Introduction to Java Programming Lecture 16 Super and Sub Classes.
:Problem E.Stone Game ★★★☆☆ 題組: Problem Set Archive with Online Judge 題號: 10165: Problem E.Stone Game 解題者:李濟宇 解題日期: 2006 年 3 月 26 日 題意: Jack 與 Jim.
幼兒行為觀察與記錄 第八章 事件取樣法.
VHDL語法(3).
1 柱體與錐體 1. 找出柱體與錐體的規則 2. 柱體的命名與特性 3. 柱體的展開圖 4. 錐體的命名與特性 5. 錐體的展開圖
ERP Purge Manual 2007/10/2 TinaLee. 2 Purge Job  Interface Purge  Transaction Purge  Other DataPurge (Like Customer 、 Vendor 、 Employee Data)
1 Chap. 7 Response of First-Order RL and RC Circuits Contents 7.1 The Natural Response of an RL Circuit 7.2 The Natural Response of an RC Circuit 7.3 The.
2018/6/26 An Energy-efficient TCAM-based Packet Classification with Decision-tree Mapping Author: Zhao Ruan, Xianfeng Li , Wenjun Li Publisher: 2013.
2019/4/9 Fast Packet Classification on OpenFlow Switches Using Multiple R*-Tree Based Bitmap Intersection Author: Ding-Fong Huang and Chien Chen Publisher/Conf.:
Large-scale Packet Classification on FPGA
Packet Classification Using Dynamically Generated Decision Trees
Fast Network Congestion Detection And Avoidance Using P4
2019/8/30 Modified ElGamal Elliptic Curve Cryptosystem using Hexadecimal Representation Author: Ziad E. Dawahdeh, Shahrul N. Yaakob and Ali Makki Sagheer.
Presentation transcript:

BiCuts: A fast packet classification algorithm using bit-level cutting 2019/7/29 BiCuts: A fast packet classification algorithm using bit-level cutting Authors :Zhi Liu,Shijie Sun,Hang Zhu,Jiaqi Gao ,Jun Presenter : Yi-Fang, Huang Conference :  Computer Communications volume109(September 2017) 1 CSIE CIAL Lab

2019/7/29 Introduction This paper proposes BitCuts, a decision-tree algorithm that performs bit-level cutting. The contributions of this paper are summarized as follows: We study the cutting schemes of existing decision-tree algorithms, and reveal their inefficiencies in terms of speed and space. “Bit-level cut” is able to zoom into densely clustered rule space and cut at the right granularity, which avoids unnecessary partitions and excessive memory consumption. BitCuts uses parallel bitindexing to support fast child-node traversal and enable large node fanout. For 5-tuple rules, the child-node indexing can be implemented by two bit-manipulation instructions, achieving ultra-fast decision-tree traversal. In order to build an efficient decision tree, a bit-selection algorithm is proposed to determine the cutting bits at each tree node, targeted at finding the most effective bits for separating the rules. 由於FPGA的可重複改寫的特性和高頻寬介面,使FPGA成為發展網路架構的選擇。 但是,它們的有限內存限制了可以存儲在芯片內部的information,並且off chip memory通常太慢以至於不能滿足高速網絡的需求。 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

2019/7/29 The Bitcuts overview 過濾幾個數據包時,所有通過link的數據包必須檢查filter為 正常運行的 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

Classification operation 2019/7/29 Classification operation 過濾幾個數據包時,所有通過link的數據包必須檢查filter為 正常運行的 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

Classification operation 2019/7/29 Classification operation 過濾幾個數據包時,所有通過link的數據包必須檢查filter為 正常運行的 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

Classification operation 2019/7/29 Classification operation 過濾幾個數據包時,所有通過link的數據包必須檢查filter為 正常運行的 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

2019/7/29 Rule set update The Micron AC-510 Board was chosen for its HMC memory module, but it lacks the network interface that would allow our system to be tested in a real network deployment. Instead, our architecture was tested with a spoofed network interface: packets were generated and results verified onchip. The Test Packet Generator creates and sends a continuous stream of packets at a line rate of up to 10 Gbps. Rather than generate random packets on-chip, this component holds a group of pre-generated packets in on-chip memory and sends them in an infinite loop. The random packets themselves are generated in software and streamed to the test framework at initialization time through the Pico Framework. Note, since the random packets are generated in software, the expected match results can also be computed in software and sent to the system at initialization time Spoofed:模仿 由軟體產生Packet且透過Test Packet Generator 把預先產生的Packet存到on-chip memory(block ram) ,然後用一個無窮迴圈重複送。 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

2019/7/29 Rule set update Spoofed:模仿 由軟體產生Packet且透過Test Packet Generator 把預先產生的Packet存到on-chip memory(block ram) ,然後用一個無窮迴圈重複送。 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

2019/7/29 Rule set update The Micron AC-510 Board was chosen for its HMC memory module, but it lacks the network interface that would allow our system to be tested in a real network deployment. Instead, our architecture was tested with a spoofed network interface: packets were generated and results verified onchip. The Test Packet Generator creates and sends a continuous stream of packets at a line rate of up to 10 Gbps. Rather than generate random packets on-chip, this component holds a group of pre-generated packets in on-chip memory and sends them in an infinite loop. The random packets themselves are generated in software and streamed to the test framework at initialization time through the Pico Framework. Note, since the random packets are generated in software, the expected match results can also be computed in software and sent to the system at initialization time Spoofed:模仿 由軟體產生Packet且透過Test Packet Generator 把預先產生的Packet存到on-chip memory(block ram) ,然後用一個無窮迴圈重複送。 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

2019/7/29 Evaluation 如果user指定了某個pad(Pcap中某個Packet段),則它將包含在模塊頭中 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

Packet Throughput & Processing Latency 2019/7/29 Packet Throughput & Processing Latency Two entries in the table correspond to our work: the system tested with 1504 rules (max rule count at 10 Gbps) and 2^20= 1048576 rules, as the first and second entries respectively. We note that previous systems outperform our implementation at lower rule counts; this is expected, since these systems utilize only on-chip memory. Our off-chip memory solution is the only hardware system (to the best of our knowledge) that can support much larger rule counts, achieving a processing rate of 16.4 Mbps at 1M rules. Table 2 presents the packet throughput of our best system, 160 systolic PEs, along with other state-of-the-art hardware implementations of packet matching systems. Two entries in the table correspond to our work: the system tested with 1504 rules (max rule count at 10 Gbps) and 220 = 1048576 rules, as the first and second entries respectively. We note that previous systems outperform our implementation at lower rule counts; this is expected, since these systems utilize only on-chip memory. Our off-chip memory solution is the only hardware system (to the best of our knowledge) that can support much larger rule counts, achieving a processing rate of 16.4 Mbps at 1M rules. National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab

2019/7/29 Resource Utilization 如果user指定了某個pad(Pcap中某個Packet段),則它將包含在模塊頭中 National Cheng Kung University CSIE Computer & Internet Architecture Lab CSIE CIAL Lab