Brocade: Landmark Routing on Overlay Networks

Slides:



Advertisements
Similar presentations
Brocade: Landmark Routing on Peer to Peer Networks Ben Y. Zhao Yitao Duan, Ling Huang, Anthony Joseph, John Kubiatowicz IPTPS, March 2002.
Advertisements

Dynamic Replica Placement for Scalable Content Delivery Yan Chen, Randy H. Katz, John D. Kubiatowicz {yanchen, randy, EECS Department.
Ion Stoica, Robert Morris, David Karger, M. Frans Kaashoek, Hari Balakrishnan MIT and Berkeley presented by Daniel Figueiredo Chord: A Scalable Peer-to-peer.
Peer to Peer and Distributed Hash Tables
Scalable Content-Addressable Network Lintao Liu
Kademlia: A Peer-to-peer Information System Based on the XOR Metric.
CHORD – peer to peer lookup protocol Shankar Karthik Vaithianathan & Aravind Sivaraman University of Central Florida.
Lecture 5 - Routing On the Flat Labels M.Sc Ilya Nikolaevskiy Helsinki Institute for Information Technology (HIIT)
Chord: A Scalable Peer-to-peer Lookup Service for Internet Applications Robert Morris Ion Stoica, David Karger, M. Frans Kaashoek, Hari Balakrishnan MIT.
Chord: A Scalable Peer-to-peer Lookup Service for Internet Applications Ion StoicaRobert Morris David Liben-NowellDavid R. Karger M. Frans KaashoekFrank.
Massively Distributed Database Systems Distributed Hash Spring 2014 Ki-Joune Li Pusan National University.
Presented by Elisavet Kozyri. A distributed application architecture that partitions tasks or work loads between peers Main actions: Find the owner of.
Looking Up Data in P2P Systems Hari Balakrishnan M.Frans Kaashoek David Karger Robert Morris Ion Stoica.
Brocade Landmark Routing on Structured P2P Overlays Ben Zhao, Yitao Duan, Ling Huang Anthony Joseph and John Kubiatowicz (IPTPS 2002) Goals Improve routing.
Overlay Networks EECS 122: Lecture 18 Department of Electrical Engineering and Computer Sciences University of California Berkeley.
Chord: A Scalable Peer-to-peer Lookup Service for Internet Applications Ion Stoica, Robert Morris, David Karger, M. Frans Kaashoek and Hari alakrishnan.
Secure Overlay Services Adam Hathcock Information Assurance Lab Auburn University.
Topics in Reliable Distributed Systems Fall Dr. Idit Keidar.
1 CS 194: Distributed Systems Distributed Hash Tables Scott Shenker and Ion Stoica Computer Science Division Department of Electrical Engineering and Computer.
Decentralized Location Services CS273 Guest Lecture April 24, 2001 Ben Y. Zhao.
Or, Providing Scalable, Decentralized Location and Routing Network Services Tapestry: Fault-tolerant Wide-area Application Infrastructure Motivation and.
3/30/2005 Auburn University Information Assurance Lab 1 Simulating Secure Overlay Services.
Data Consistency in the Structured Peer-to-Peer Network Cheng-Ying Ou, Polly Huang Network and Systems Lab 台灣大學電機資訊學院電機所.
Security Considerations for Structured p2p Peng Wang 6/04/2003.
By Sylvia Ratnasamy, Andrey Ermolinskiy, Scott Shenker Presented by Fei Jia Revisiting IP Multicast.
IDRM: Inter-Domain Routing Protocol for Mobile Ad Hoc Networks C.-K. Chau, J. Crowcroft, K.-W. Lee, S. H.Y. Wong.
Brocade Landmark Routing on P2P Networks Gisik Kwon April 9, 2002.
Vincent Matossian September 21st 2001 ECE 579 An Overview of Decentralized Discovery mechanisms.
Ion Stoica, Robert Morris, David Karger, M. Frans Kaashoek, Hari Balakrishnan MIT and Berkeley presented by Daniel Figueiredo Chord: A Scalable Peer-to-peer.
TOMA: A Viable Solution for Large- Scale Multicast Service Support Li Lao, Jun-Hong Cui, and Mario Gerla UCLA and University of Connecticut Networking.
Peer-to-Peer Supported Cache System for File Transfer Joonbok Lee
1 Peer-to-Peer Technologies Seminar by: Kunal Goswami (05IT6006) School of Information Technology Guided by: Prof. C.R.Mandal, School of Information Technology.
Peer to Peer A Survey and comparison of peer-to-peer overlay network schemes And so on… Chulhyun Park
15 November 2005LCN Collision Detection and Resolution in Hierarchical Peer-to-Peer Systems Verdi March 1, Yong Meng Teo 1,2, Hock Beng Lim 2, Peter.
1 Secure Peer-to-Peer File Sharing Frans Kaashoek, David Karger, Robert Morris, Ion Stoica, Hari Balakrishnan MIT Laboratory.
Peer to Peer Network Design Discovery and Routing algorithms
Tapestry : An Infrastructure for Fault-tolerant Wide-area Location and Routing Presenter : Lee Youn Do Oct 5, 2005 Ben Y.Zhao, John Kubiatowicz, and Anthony.
LOOKING UP DATA IN P2P SYSTEMS Hari Balakrishnan M. Frans Kaashoek David Karger Robert Morris Ion Stoica MIT LCS.
CS 347Notes081 CS 347: Parallel and Distributed Data Management Notes 08: P2P Systems.
Constructing Inter-Domain Packet Filters to Control IP Spoofing Based on BGP Updates Zhenhai Duan, Xin Yuan Department of Computer Science Florida State.
Incrementally Improving Lookup Latency in Distributed Hash Table Systems Hui Zhang 1, Ashish Goel 2, Ramesh Govindan 1 1 University of Southern California.
Fall, 2001CS 6401 Switching and Routing Outline Routing overview Store-and-Forward switches Virtual circuits vs. Datagram switching.
A Survey of Peer-to-Peer Content Distribution Technologies Stephanos Androutsellis-Theotokis and Diomidis Spinellis ACM Computing Surveys, December 2004.
Simple Load Balancing for Distributed Hash tables
A Cluster-based Routing Protocol for Mobile Ad hoc Networks
Ion Stoica, Robert Morris, David Liben-Nowell, David R. Karger, M
Magdalena Balazinska, Hari Balakrishnan, and David Karger
DELAY TOLERANT NETWORK
A Scalable Peer-to-peer Lookup Service for Internet Applications
Impact of Neighbor Selection on Performance and Resilience of Structured P2P Networks Sushma Maramreddy.
What Are Routers? Routers are an intermediate system at the network layer that is used to connect networks together based on a common network layer protocol.
Plethora: Infrastructure and System Design
Zhichen Xu, Mallik Mahalingam, Magnus Karlsson
Turning Heterogeneity into an Advantage in Overlay Routing
Early Measurements of a Cluster-based Architecture for P2P Systems
EE 122: Peer-to-Peer (P2P) Networks
Overlay Networking Overview.
Building Peer-to-Peer Systems with Chord, a Distributed Lookup Service
Internet Indirection Infrastructure
Distributed Publish/Subscribe Network
Distributed Hash Tables
Rapid Mobility via Type Indirection
Small-Scale Peer-to-Peer Publish/Subscribe
Dynamic Replica Placement for Scalable Content Delivery
EE 122: Lecture 22 (Overlay Networks)
DNSR: Domain Name Suffix-based Routing in Overlay Networks
MIT LCS Proceedings of the 2001 ACM SIGCOMM Conference
Consistent Hashing and Distributed Hash Table
A. D. Sotiriou, P. Kalliaras, N. Mitrou
Brocade: Landmark Routing on Peer to Peer Networks
Presentation transcript:

Brocade: Landmark Routing on Overlay Networks To P2P or not to P2P? http://www.cs.berkeley.edu/~duan/prjs/cs262/ CS262A Fall 2001 Yitao Duan and Ling Huang duan@cs.berkeley.edu, hlion@newton.berkeley.edu

Motivation Problems with existing P2P Network Constrained by the theoretical approach adopted, nodes are treated uniformly[1, 2, 3, 4] Routing algorithms are decoupled from underlying topology and node capability Result: inefficient routing Reality: Nodes are not born equal Bandwidth, Connectivity, Storage, Processing Power. Administrative Constraints

Brocade: Discrimination Justified A philosophy: A system is more efficient when it is organized – e.g., IP routing on Internet Respect the differences and take advantage of those that are more powerful – Supernodes! Fast/well-connected/situated near network access points Supernodes have better knowledge of underlying network characteristics. Benefit from aggregation. Construct a hierarchy out of flat network

Brocade Architecture Brocade P2P Network S D Original Route AS-2 P2P Network AS-1 AS-3 Brocade S D Original Route Brocade Route

Message filtering: only send inter-domain messages to Brocade. Overlay nodes are grouped by their supernodes – Cover Set Supernodes treat their overlay nodes as objects that they possess Routing on Brocade => Object Location. Use your favorite mechanism: Tapestry[1], CAN[3], Chord[2], Pastry[4] … Message filtering: only send inter-domain messages to Brocade.

Case Study - Brocade On Tapestry Tapestry: A novel wide-area fault-tolerant location and routing infrastructure[1] Construction Gateway routers or machines close by as supernodes Existing connections among supernodes as Brocade links Routing: object location Tapestry style Each supernode advertises the IDs of overlay nodes in its cover set as IDs of objects it “stores”. Destination’s supernode can be found using Tapestry’s object location mechanism Remaining issue: How to get onto Brocade?

Get onto the Super Highway Naïve Brocade: Tapestry routing unchanged. Message gets onto the Brocade overlay if a supernode is encountered on its route. Advantage: simple, no modification to ordinary nodes. Disadvantage: possibility of hitting a supernode in Tapestry routing small. IP Snooping Brocade: Supernodes snoop IP packets to intercept Tapestry messages. Advantage: No modification to ordinary nodes. High possibility of encountering supernodes because supernodes are situated near the edge of local networks. Disadvantage: Difficult to implement

Directed Brocade: Each overlay node keep info about its supernode and decides by its own whether to send a message to supernode directly. Feasible: only local information required Decision Engine: A small cache storing most frequently used nodes in its cover set will do the trick. Query locality will make hit rate high Consequences of mistakes aren’t expensive Destination is in my cover set? Send to supernode Ordinary Tapestry Routing No Yes

Fig 1. Hops Based RDP Fig 2. Aggregate bandwidth used per message Simulation Results Fig 1. Hops Based RDP Fig 2. Aggregate bandwidth used per message

Optimizing Object Location on Brocade Routing latency could be high if latencies on Brocade links are high and object location on Brocade is not optimized(Fig 3) Optimization: Bloom Filter - Membership query and group ID problem Fig 3. Weighted latency RDP w/o optimization Fig 4. Weighted latency RDP with Bloom Filter Brocade link latency/Ordinary link latency = 8 : 1

Conclusion and Future work Brocade: powerful idea that can achieve near optimal performance General enough to be applied to other (P2P) networks Future research: Study the effect of different supernodes selection and distribution Further optimization of object location on Brocade overlay Latent Brocade Brocade benefits from aggregation of info Bias some nodes in the network so they will be favored by others while selecting route - an implicit Brocade

References [1] ZHAO, B. Y., KUBIATOWICZ, J. D., AND JOSEPH, A. D. Tapestry: An infrastructure for fault-tolerant wide-area location and routing. Tech. Rep. UCB/CSD-01-1141, University of California at Berkeley, Computer Science Division, April 2001. [2] STOICA, I., MORRIS, R., KARGER, D., KAASHOEK, M. F., AND BALAKRISHNAN, H. Chord: A scalable peer-to-peer lookup service for internet applications. In Proceedings of SIGCOMM (August2001), ACM. [3] RATNASAMY, S., FRANCIS, P., HANDLEY, M., KARP, R., AND SCHENKER, S. A scalable content-addressable network. In Proceedings of SIGCOMM (August 2001), ACM. [4] ROWSTRON, A., AND DRUSCHEL, P. Pastry: Scalable, distributed object location and routing for large-scale peer-to-peer systems. In Proceedings of IFIP/ACM Middleware 2001 (November 2001). [5] TSUCHIYA, P. F. The landmark hierarchy: A new hierarchy for routing in very large networks. Computer Communication Review 18, 4 (August 1988), 35–42.