1 OpenFlow + : Extension for OpenFlow and its Implementation Hongyu Hu, Jun Bi, Tao Feng, You Wang, Pingping Lin Tsinghua University 2011-08-12.

Slides:



Advertisements
Similar presentations
TWO STEP EQUATIONS 1. SOLVE FOR X 2. DO THE ADDITION STEP FIRST
Advertisements

Requirements Engineering Processes – 2
MCT620 – Distributed Systems
Computer Networks TCP/IP Protocol Suite.
1 UNIT I (Contd..) High-Speed LANs. 2 Introduction Fast Ethernet and Gigabit Ethernet Fast Ethernet and Gigabit Ethernet Fibre Channel Fibre Channel High-speed.
Copyright © 2003 Pearson Education, Inc. Slide 1 Computer Systems Organization & Architecture Chapters 8-12 John D. Carpinelli.
Chapter 1 The Study of Body Function Image PowerPoint
1 Copyright © 2013 Elsevier Inc. All rights reserved. Chapter 1 Embedded Computing.
Copyright © 2011, Elsevier Inc. All rights reserved. Chapter 6 Author: Julia Richards and R. Scott Hawley.
Author: Julia Richards and R. Scott Hawley
1 Copyright © 2013 Elsevier Inc. All rights reserved. Appendix 01.
OSPF 1.
FI Research in China Jun Bi Tsinghua Univ./CERNET Beijing China.
1 Building a Fast, Virtualized Data Plane with Programmable Hardware Bilal Anwer Nick Feamster.
UNITED NATIONS Shipment Details Report – January 2006.
1 Hyades Command Routing Message flow and data translation.
18 Copyright © 2005, Oracle. All rights reserved. Distributing Modular Applications: Introduction to Web Services.
1 RA I Sub-Regional Training Seminar on CLIMAT&CLIMAT TEMP Reporting Casablanca, Morocco, 20 – 22 December 2005 Status of observing programmes in RA I.
Jeopardy Q 1 Q 6 Q 11 Q 16 Q 21 Q 2 Q 7 Q 12 Q 17 Q 22 Q 3 Q 8 Q 13
Jeopardy Q 1 Q 6 Q 11 Q 16 Q 21 Q 2 Q 7 Q 12 Q 17 Q 22 Q 3 Q 8 Q 13
FACTORING ax2 + bx + c Think “unfoil” Work down, Show all steps.
Year 6 mental test 5 second questions
Year 6 mental test 10 second questions
|epcc| NeSC Workshop Open Issues in Grid Scheduling Ali Anjomshoaa EPCC, University of Edinburgh Tuesday, 21 October 2003 Overview of a Grid Scheduling.
ZMQS ZMQS
Communicating over the Network
INTERNET PROTOCOLS Class 9 CSCI 6433 David C. Roberts Entire contents copyright 2011, David C. Roberts, all rights reserved.
Solve Multi-step Equations
REVIEW: Arthropod ID. 1. Name the subphylum. 2. Name the subphylum. 3. Name the order.
DOROTHY Design Of customeR dRiven shOes and multi-siTe factorY Product and Production Configuration Method (PPCM) ICE 2009 IMS Workshops Dorothy Parallel.
Chapter 1: Introduction to Scaling Networks
TCP/IP Protocol Suite 1 Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display. Chapter 2 The OSI Model and the TCP/IP.
© 2006 Cisco Systems, Inc. All rights reserved. MPLS v MPLS VPN Technology Introducing MPLS VPN Architecture.
Chapter 15 Integrated Services Digital Network ISDN Services History Subscriber Access Layers BISDN WCB/McGraw-Hill The McGraw-Hill Companies, Inc., 1998.
2 |SharePoint Saturday New York City
IP Multicast Information management 2 Groep T Leuven – Information department 2/14 Agenda •Why IP Multicast ? •Multicast fundamentals •Intradomain.
VOORBLAD.
15. Oktober Oktober Oktober 2012.
Chapter 20 Network Layer: Internet Protocol
Copyright © 2013, 2009, 2006 Pearson Education, Inc.
Factor P 16 8(8-5ab) 4(d² + 4) 3rs(2r – s) 15cd(1 + 2cd) 8(4a² + 3b²)
Squares and Square Root WALK. Solve each problem REVIEW:
Basel-ICU-Journal Challenge18/20/ Basel-ICU-Journal Challenge8/20/2014.
1..
© 2012 National Heart Foundation of Australia. Slide 2.
Understanding Generalist Practice, 5e, Kirst-Ashman/Hull
Executional Architecture
Global Analysis and Distributed Systems Software Architecture Lecture # 5-6.
SIMOCODE-DP Software.
25 seconds left…...
H to shape fully developed personality to shape fully developed personality for successful application in life for successful.
Januar MDMDFSSMDMDFSSS
1 © 2003, Cisco Systems, Inc. All rights reserved. CCNA TCP/IP Protocol Suite and IP Addressing Halmstad University Olga Torstensson
Chapter 10: The Traditional Approach to Design
Analyzing Genes and Genomes
Systems Analysis and Design in a Changing World, Fifth Edition
We will resume in: 25 Minutes.
© 2007 Cisco Systems, Inc. All rights reserved.Cisco Public 1 Addressing the Network – IPv4 Network Fundamentals – Chapter 6.
©Brooks/Cole, 2001 Chapter 12 Derived Types-- Enumerated, Structure and Union.
Intracellular Compartments and Transport
PSSA Preparation.
VPN AND REMOTE ACCESS Mohammad S. Hasan 1 VPN and Remote Access.
Essential Cell Biology
14.1 Chapter 14 Wireless LANs Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
McGraw-Hill©The McGraw-Hill Companies, Inc., 2001 Chapter 16 Integrated Services Digital Network (ISDN)
TCP/IP Protocol Suite 1 Chapter 18 Upon completion you will be able to: Remote Login: Telnet Understand how TELNET works Understand the role of NVT in.
Presentation transcript:

1 OpenFlow + : Extension for OpenFlow and its Implementation Hongyu Hu, Jun Bi, Tao Feng, You Wang, Pingping Lin Tsinghua University

2 Outline Problems about current Internet What OpenFlow brings to us Some aspects need to be Improved in OpenFlow OpenFlow + : four extensions for OpenFlow OpenFlow + s implementation Benefits of OpenFlow + Applications of OpenFlow + Conclusions and future work

3 Problems about current Internet Internet has made great success and big progress. However, the network-layer of Internet and the network devices in Internet have been relatively stagnant. Few changes or improvements have been made in last forty years, which is a stark contrast to the prosperity of the application-layer of Internet. In our opnion, all of these is mainly due to the lack of openness in the network-layer.

4 What OpenFlow brings to us OpenFlow aims to enable innovation for the network-layer and network devices. Valuable thoughts that OpenFlow brought to us: –1) The design of FlowTable in the data plane realized the standardization, simplification and openness of network hardware. –2) The design of OpenFlow protocol realized the standardization and openness of the access interfaces to network hardware. –3) User-defined control logics can be easily added to the Controller as new components. –4) The centralized computing mode designed in OpenFlow makes some functions or services based on global information possible.

5 What OpenFlow brings to us

6 Some aspects need to be Improved in OpenFlow Standard hardware in OpenFlow needs to be extended. FlowTable hardware in OpenFlow can be realized low-costly and quickly. Control mode needs to be extended. Communication interface needs to be extended.

7 OpenFlow + : four extensions for OpenFlow We proposed some extensions for OpenFlow (OpenFlow + ) in this paper: –Standard hardware extension –Control mode extension –Communication interface extension –Low-cost FlowTable realization

8 Extensions 1 - More Hardwares More standard hardware resources need to be exposed and extended –Only FlowTable hardware is not enough –Other hardwares become mature –Little difference of these hardware in different venders devices ACL&QoS, FIB, and Sample hardwares are exposed for outside control logic

9 Extensions 1 - More Hardwares

10 Extensions 2 - More Control Modes A coexisting collaborative mode of distributed computing and centralized computing is needed. –The pure external centralized control mode is deficient in control efficiency and function maturity. –The pure internal distributed control mode is deficient in global coordinate computing.

11 Extensions 2 - More Control Modes

12 Extensions 2 - More Control Modes Rules to confirm both types of computing to work together and collaborate smoothly are designed: –Both type of control logics can exchange data information with each other. –Both type of control logics have the same abilities to control the open hardware resources. –If control conflicts occurs to same hardware resources, we designed some rules to solve these conflicts.

13 Extensions 2 - More Control Modes Rules to resolve control conflicts: –Establish an arbitration module to execute the optimal selection for both controls from the inside and outside. –Use fixed priority levels to choose the optimal control. –Both of the control operations will coexist.

Extensions 3-Data reorganized by TLV To support more types of information exchange between the control plane and the data plane, and to support the easy extension of the length of existing information in the OpenFlow protocol, we introduce TLV (Type Length Value, TLV) format to the OpenFlow protocol to reorganize the information in it. 14

Extensions 3-Data reorganized by TLV TLV format can: –Efficiently organize data with variable length –Conveniently implement the extension for the length and type of data In TLV format, each piece of data is organized by the triple of (Type, Length, Value) TLV can be used or arranged recursively 15

Extensions 3-Data reorganized by TLV 16 Table 1. TLV general format.

Extensions 3-Data reorganized by TLV 17 It will become very easy for us to extend IPv4 address format to IPv6 address format if the TLV format is used in the organization of FlowTable.

18 Extensions 4 – Using existing hardwares to realize FlowTable Many mature hardware resources inside devices can be utilized to implement the base functions of FlowTable, such as ACL&QoS hardware tables and FIB hardware tables. When using the combination of ACL&QoS and FIB hardware to implement FlowTable, two different types of FlowTable are needed: –ACL&QoS-type FlowTable –FIB-type FlowTable

Extensions 4 – Using existing hardwares to realize FlowTable On the FlowTable Sender side: –FlowTable needs to be organized and described by two kinds of TLVs. –Different values of Type field in TLVs identify different types of FlowTable. –Any type of FlowTable contains three sub- TLVs: Header sub-TLV, Action sub-TLV, and Counter sub-TLV(optional). –In different types of FlowTable, the specific content of these three sub-TLVs may be different. 19

20 Extensions 4 – Using existing hardwares to realize FlowTable

21 Extensions 4 – Using existing hardwares to realize FlowTable

22 Extensions 4 – Using existing hardwares to realize FlowTable On the FlowTable reseiver side, the FlowTable receivers will: –Analyze the Typefield of the TLVs received, –Distinguish ACL&QoS-type and FIB-type FlowTables according to the TLV type –Issue a corresponding item to ACL&QoS or to FIB hardware tables –OpenFlow Agent Module will execute the change from FlowTable TLV to ACL&QoS or FIB hardware resources.

23 OpenFlow + s implementation We implemented OpenFlow + in a commercial router (DCRS 5980/5950, DigitalChina Company, RouterSwitch) -- OpenRouter. There is no FlowTalbe hardware in OpenRouters. FlowTable is replaced by ACL and FIB.

24 OpenFlow + s implementation OpenRouter architecture

25 OpenFlow + s implementation The details of OpenRouter: –An OpenFlow Agent module is embedded into the control software of OpenRouter. –OpenFlow protocol is redesigned and reconstructed with TLV. –FlowTable is implemented using existing hardware resources -- ACL&QoS and FIB. –OpenFlow Agent module acts as an interfaces with Routing Table Management (RTM) module and sFlow module. –Two asynchronous messages and a synchronous message are added to transmit FIB and sampling packets.

26 Benefits of OpenFlow + More openness for network devices. More efficient control for the network. More flexible organization for data in OpenFlow Protocol. More low-cost implementation for OpenFlow hardware.

27 Applications of OpenFlow + (1) Intra-AS Source Address Validation

28 Applications of OpenFlow + (2) QoS routing in intra-AS

29 Conclusions and Future Work Internet needs innovation. But we still dont know exactly what functions and features that the future Internet should include. We think, it may be not proper to build a concrete and fixed network for the future Internet now. We think, innovation ability is what the future Internet really needs.

30 Conclusions and Future Work OpenFlows openness and standardization give Internet more powerful abilities to reform and innovate. In this paper, we made four extensions for OpenFlow and realized these extensions in a type of vender devices. Applications about how to use these extensions are introduced.

31 Conclusions and Future Work More hardware resources in devices should be exposed and standardized The methods for inside or outside control logics to use these open and standard hardware resources should be unified

32 Thank you! Q&A