HybridCuts: A Scheme Combining Decomposition and Cutting for Packet Classification Author: Wenjun Li, Xianfeng Li Publisher: 2013 IEEE 21 st Annual Symposium.

Slides:



Advertisements
Similar presentations
Packet Classification using Hierarchical Intelligent Cuttings
Advertisements

Multi-dimensional Packet Classification on FPGA: 100Gbps and Beyond
1 IP-Lookup and Packet Classification Advanced Algorithms & Data Structures Lecture Theme 08 – Part I Prof. Dr. Th. Ottmann Summer Semester 2006.
Balajee Vamanan, Gwendolyn Voskuilen, and T. N. Vijaykumar School of Electrical & Computer Engineering SIGCOMM 2010.
Scalable Packet Classification Using Hybrid and Dynamic Cuttings Authors : Wenjun Li,Xianfeng Li Publisher : Engineering Lab on Intelligent Perception.
Fast Firewall Implementation for Software and Hardware-based Routers Lili Qiu, Microsoft Research George Varghese, UCSD Subhash Suri, UCSB 9 th International.
Ultra-High Throughput Low-Power Packet Classification
A Ternary Unification Framework for Optimizing TCAM-Based Packet Classification Systems Author: Eric Norige, Alex X. Liu, and Eric Torng Publisher: ANCS.
1 TCAM Razor: A Systematic Approach Towards Minimizing Packet Classifiers in TCAMs Department of Computer Science and Information Engineering National.
IP Address Lookup for Internet Routers Using Balanced Binary Search with Prefix Vector Author: Hyesook Lim, Hyeong-gee Kim, Changhoon Publisher: IEEE TRANSACTIONS.
Firewall Policy Queries Author: Alex X. Liu, Mohamed G. Gouda Publisher: IEEE Transaction on Parallel and Distributed Systems 2009 Presenter: Chen-Yu Chang.
1 On Constructing Efficient Shared Decision Trees for Multiple Packet Filters Author: Bo Zhang T. S. Eugene Ng Publisher: IEEE INFOCOM 2010 Presenter:
1 Memory-Efficient 5D Packet Classification At 40 Gbps Authors: Ioannis Papaefstathiou, and Vassilis Papaefstathiou Publisher: IEEE INFOCOM 2007 Presenter:
1 Energy Efficient Multi-match Packet Classification with TCAM Fang Yu
Efficient Multidimensional Packet Classification with Fast Updates Author: Yeim-Kuan Chang Publisher: IEEE TRANSACTIONS ON COMPUTERS, VOL. 58, NO. 4, APRIL.
Parallel-Search Trie-based Scheme for Fast IP Lookup
Performance Evaluation of IPv6 Packet Classification with Caching Author: Kai-Yuan Ho, Yaw-Chung Chen Publisher: ChinaCom 2008 Presenter: Chen-Yu Chaug.
1 Energy Efficient Packet Classification Hardware Accelerator Alan Kennedy, Xiaojun Wang HDL Lab, School of Electronic Engineering, Dublin City University.
Study of IP address lookup Schemes
Packet Classification George Varghese. Original Motivation: Firewalls Firewalls use packet filtering to block say ssh and force access to web and mail.
1 EffiCuts : Optimizing Packet Classification for Memory and Throughput Author: Balajee Vamanan, Gwendolyn Voskuilen and T. N. Vijaykumar Publisher: ACM.
HybridCuts: A Scheme Combining Decomposition and Cutting for Packet Classification Wenjun Li Xianfeng Li School of Electronic and Computer Engineering.
March 1, Packet Classification and Filtering for Network Processors JC Ho.
1 Efficient packet classification using TCAMs Authors: Derek Pao, Yiu Keung Li and Peng Zhou Publisher: Computer Networks 2006 Present: Chen-Yu Lin Date:
High-Performance Packet Classification on GPU Author: Shijie Zhou, Shreyas G. Singapura and Viktor K. Prasanna Publisher: HPEC 2014 Presenter: Gang Chi.
HybridCuts: A Scheme Combining Decomposition and Cutting for Packet Classification Authors : Wenjun Li, Xianfeng Li Publisher : 2013 IEEE 21st Annual Symposium.
Packet Classification using Rule Caching Author: Nitesh B. Guinde, Roberto Rojas-Cessa, Sotirios G. Ziavras Publisher: IISA, 2013 Fourth International.
Packet Classification Using Multi-Iteration RFC Author: Chun-Hui Tsai, Hung-Mao Chu, Pi-Chung Wang Publisher: COMPSACW, 2013 IEEE 37th Annual (Computer.
(TPDS) A Scalable and Modular Architecture for High-Performance Packet Classification Authors: Thilan Ganegedara, Weirong Jiang, and Viktor K. Prasanna.
An Improved Algorithm to Accelerate Regular Expression Evaluation Author: Michela Becchi, Patrick Crowley Publisher: 3rd ACM/IEEE Symposium on Architecture.
Multi-dimensional Packet Classification on FPGA 100 Gbps and Beyond Author: Yaxuan Qi, Jeffrey Fong, Weirong Jiang, Bo Xu, Jun Li, Viktor Prasanna Publisher:
Wire Speed Packet Classification Without TCAMs ACM SIGMETRICS 2007 Qunfeng Dong (University of Wisconsin-Madison) Suman Banerjee (University of Wisconsin-Madison)
Fast Packet Classification Using Bloom filters Authors: Sarang Dharmapurikar, Haoyu Song, Jonathan Turner, and John Lockwood Publisher: ANCS 2006 Present:
Packet Classification on Multiple Fields 참고 논문 : Pankaj Gupta and Nick McKeown SigComm 1999.
Multi-Field Range Encoding for Packet Classification in TCAM Author: Yeim-Kuan Chang, Chun-I Lee and Cheng-Chien Su Publisher: INFOCOM 2011 Presenter:
Balajee Vamanan and T. N. Vijaykumar School of Electrical & Computer Engineering CoNEXT 2011.
Author : Guangdeng Liao, Heeyeol Yu, Laxmi Bhuyan Publisher : Publisher : DAC'10 Presenter : Jo-Ning Yu Date : 2010/10/06.
Scalable Many-field Packet Classification on Multi-core Processors Authors : Yun R. Qu, Shijie Zhou, Viktor K. Prasanna Publisher : International Symposium.
StrideBV: Single chip 400G+ packet classification Author: Thilan Ganegedara, Viktor K. Prasanna Publisher: HPSR 2012 Presenter: Chun-Sheng Hsueh Date:
1 Fast packet classification for two-dimensional conflict-free filters Department of Computer Science and Information Engineering National Cheng Kung University,
PARALLEL-SEARCH TRIE- BASED SCHEME FOR FAST IP LOOKUP Author: Roberto Rojas-Cessa, Lakshmi Ramesh, Ziqian Dong, Lin Cai Nirwan Ansari Publisher: IEEE GLOBECOM.
IPv6-Oriented 4 OC768 Packet Classification with Deriving-Merging Partition and Field- Variable Encoding Scheme Mr. Xin Zhang Undergrad. in Tsinghua University,
A Smart Pre-Classifier to Reduce Power Consumption of TCAMs for Multi-dimensional Packet Classification Yadi Ma, Suman Banerjee University of Wisconsin-Madison.
TCAM –BASED REGULAR EXPRESSION MATCHING SOLUTION IN NETWORK Phase-I Review Supervised By, Presented By, MRS. SHARMILA,M.E., M.ARULMOZHI, AP/CSE.
High-Speed Policy-Based Packet Forwarding Using Efficient Multi-dimensional Range Matching Lakshman and Stiliadis ACM SIGCOMM 98.
Bit Weaving: A Non-Prefix Approach to Compressing Packet Classifiers in TCAMs Author: Chad R. Meiners, Alex X. Liu, and Eric Torng Publisher: 2012 IEEE/ACM.
Cross-Product Packet Classification in GNIFS based on Non-overlapping Areas and Equivalence Class Author: Mohua Zhang, Ge Li Publisher: AISS 2012 Presenter:
Binary-tree-based high speed packet classification system on FPGA Author: Jingjiao Li*, Yong Chen*, Cholman HO**, Zhenlin Lu* Publisher: 2013 ICOIN Presenter:
Lightweight Traffic-Aware Packet Classification for Continuous Operation Author: Shariful Hasan Shaikot, Min Sik Kim Presenter: Yen-Chun Tseng Date: 2014/11/26.
1 Bit Weaving: A Non-Prefix Approach to Compressing Packet Classifiers in TCAMs Author: Chad R. Meiners, Alex X. Liu, and Eric Torng Publisher: IEEE/ACM.
Packet Classification Using Multidimensional Cutting Sumeet Singh (UCSD) Florin Baboescu (UCSD) George Varghese (UCSD) Jia Wang (AT&T Labs-Research) Reviewed.
Packet Classification Using Dynamically Generated Decision Trees
Author: Weirong Jiang and Viktor K. Prasanna Publisher: The 18th International Conference on Computer Communications and Networks (ICCCN 2009) Presenter:
Ultra-High Throughput Low-Power Packet Classification Author: Alan Kennedy and Xiaojun Wang Accepted by IEEE Transactions on VLSI.
A Fast and Scalable IPv6 Packet Classification Author: Xiaoju Zhou, Xiaohong Huang, Qiong Sun, Wei Yang, Yan Ma Publisher: Network Infrastructure and Digital.
1 Space-Efficient TCAM-based Classification Using Gray Coding Authors: Anat Bremler-Barr and Danny Hendler Publisher: IEEE INFOCOM 2007 Present: Chen-Yu.
Hierarchical packet classification using a Bloom filter and rule-priority tries Source : Computer Communications Authors : A. G. Alagu Priya 、 Hyesook.
Author : Lynn Choi, Hyogon Kim, Sunil Kim, Moon Hae Kim Publisher/Conf : IEEE/ACM TRANSACTIONS ON NETWORKING Speaker : De yu Chen Data :
Author: Weirong Jiang, Viktor K. Prasanna Publisher: th IEEE International Conference on Application-specific Systems, Architectures and Processors.
Authors : Baohua Yang, Jeffrey Fong, Weirong Jiang, Yibo Xue, and Jun Li. Publisher : IEEE TRANSACTIONS ON COMPUTERS Presenter : Chai-Yi Chu Date.
Packet Classification Using Multi- Iteration RFC Author: Chun-Hui Tsai, Hung-Mao Chu, Pi-Chung Wang Publisher: 2013 IEEE 37th Annual Computer Software.
Hierarchical Hybrid Search Structure for High Performance Packet Classification Authors : O˜guzhan Erdem, Hoang Le, Viktor K. Prasanna Publisher : INFOCOM,
Scalable Multi-match Packet Classification Using TCAM and SRAM Author: Yu-Chieh Cheng, Pi-Chung Wang Publisher: IEEE Transactions on Computers (2015) Presenter:
2018/6/26 An Energy-efficient TCAM-based Packet Classification with Decision-tree Mapping Author: Zhao Ruan, Xianfeng Li , Wenjun Li Publisher: 2013.
Toward Advocacy-Free Evaluation of Packet Classification Algorithms
Transport Layer Systems Packet Classification
Scalable Multi-Match Packet Classification Using TCAM and SRAM
Publisher : TRANSACTIONS ON NETWORKING Author : Haoyu Song, Jonathan S
Worst-Case TCAM Rule Expansion
MEET-IP Memory and Energy Efficient TCAM-based IP Lookup
Presentation transcript:

HybridCuts: A Scheme Combining Decomposition and Cutting for Packet Classification Author: Wenjun Li, Xianfeng Li Publisher: 2013 IEEE 21 st Annual Symposium Presenter: Ching Hsuan Shih Date: 2013/12/11

Outline I. Introduction II. Background III. Related Work IV. HybridCuts V. Experimental Results

I. Introduction Two major approaches to packet classification: Architectural: TCAM Advantage: process packets at line speed Disadvantage: expensive, high power consumption, rule duplication Algorithmic: Decision-tree: Hicuts, HyperCuts, and EffiCuts Decomposition: BV

I. Introduction (Cont.) Proposed scheme: HybridCuts: a combination of decomposition and decision-tree techniques A rule set decomposition algorithm based on the observation that most rules have at least one small field. A novel one-dimensional cutting algorithm called FiCuts. A two-stage cutting framework which combines one-dimensional cutting and multi-dimensional cutting techniques.

II. Background A. The Packet Classification Problem To find a matching rule from a packet classifier for a packet. A classifier is a set of rules, with each rule R consisting of a tuple of F field values, and an action to be taken in case of a match.

II. Background (Cont.) B. Complexity in Theory From a geometric point of view, for N non-overlapping hyper- rectangles in F-dimensional space: The best bounds for locating a point are either Θ (log N) time with Θ (N F ) space, or Θ (log F-1 N) time with Θ (N) space. Impractical: with just 1000 rules and 4 fields, a solution is either impracticably (N F is about 1000G) or too slow (log F-1 N is about 1000 memory accesses).

II. Background (Cont.) C. Complexity in Practice It is infeasible to design a single algorithm that can perform well in all cases. Packet classification rules in real-life have some inherent characteristics that can be exploited to reduce the complexity. The protocol field is restricted to a small set of values. Rules specify a limited number of distinct transport port ranges. The number of address prefixes matching a given address is typically five or less. The number of rules matching a given packet is typically five or less. Many different rules share the same field values.

III. Related Work A. Parallel Bit-Vector (BV) One of most representative decomposition based solutions. It works on the individual fields of rules independently for partially matching results. Then a bit-wise AND operation on all bit vectors is performed to get the final result.

III. Related Work (Cont.)

B. HiCuts and HyperCuts Take a geometric view of the packet classification problem. HiCuts chooses one dimension to cut at one time. HyperCuts chooses multiple dimensions to cut at one time.

III. Related Work (Cont.) C. EffiCuts Separate rule set into several subsets, depending on whether the value of each dimension is wildcard (or almost wildcard). Each subset creates its own decision-tree independently using HyperCuts.

III. Related Work (Cont.) D. ParaSplit Employs a complex heuristic for rule set partitioning. It is different from EffiCuts in that its objective is for efficient hardware implementation using FPGAs.

IV. HybridCuts A. Decomposition-based Framework Definitions: Given an N-dimensional rule R = (F 1, F 2, F 3, …, F N ), Len i represents the length of field F i, and a threshold value vector T = (T 1, T 2, T 3, …, T N ). We call F i is a small field if Len i ≦ T i. Then we define the following concepts for R: Big rule: ∀ i ∈ {1, 2, 3... N}, F i in R is a big field Small rule: ∃ i ∈ {1, 2, 3... N}, F i in R is a small field

IV. HybridCuts (Cont.) We decompose a 5-tuple rule set into the following five subsets without duplicates among each other: 1. Big-subset: SA, DA, SP and DP are all big field 2. SA-subset: SA is a small field for each rule 3. DA-subset: DA is a small field for each rule 4. SP-subset: SP is a small field for each rule 2. DP-subset: DP is a small field for each rule When processing a rule R with two small fields: SA and DA, suppose the sizes of the SA-subset and the DA-subset up to now are N 1 and N 2 respectively, then rule R will go to SA-subset if N 1 ≦ N 2, or to DA-subset vice versa.

IV. HybridCuts (Cont.)

B. FiCuts: A One-Dimensional Cutting Technique Simplicity: FiCuts conducts cuttings on the subset along a fixed dimension Adaptivity: FiCuts can decide when to stop FiCuts and resort to other more effective cutting methods How to decide np: spfac * number of rules at node r ≥ ∑ number of rules at each child of node r + np

IV. HybridCuts (Cont.) binth = 4 spfac = 2

IV. HybridCuts (Cont.) C. Multi-dimensional Cutting With the shrinking of the search space, rule replications begin to rise with fine cuts. FiCuts makes the decision. If the number of cuts np at a node is less than a predefined MAXCUTS, FiCuts resorts to HyperCuts

IV. HybridCuts (Cont.) binth = 2 spfac = 2

IV. HybridCuts (Cont.) D. Optimization Most rules have at least one small IP address field (SA or DA). Based on this observation, we reduce the number of decision trees. For a 5-tuple rule set, we decompose it into three subsets: 1. Big-subset: Both SA and DA are big field 2. SA-subset: SA is a small field for each rule 3. DA-subset: DA is a small field for each rule

IV. HybridCuts (Cont.) Depends on SA, DA, SP, and DPDepends on SA, and DA

V. Experimental Results A. Memory Consumption

V. Experimental Results (Cont.) A. Memory Accesses

V. Experimental Results (Cont.) C. Potential of Parallelization It can be seen that in most cases, the worst-case height of a single tree is half of or less than the overall number of memory accesses on all the trees. This means that the trees constructed are balanced in height among each other, amenable to a parallel implementation with a potential 2x speedup.