Peer to Peer (1)

References

Chapter 2.9 of Kurose and Ross Papers

o OpenNap: Open Source Napster Server o J. Liang, R. Kumar and K. Ross, Understanding KaZaA

Acknowledgements: Many of the figures are from other presentations especially from the original authors.

Client-Server Model

Let’s look at the Client-Server modelServers are centrally maintained and administeredClient has fewer computing resources than a serverThis is the way the web worksNo interaction between clients

Client Server Model

Disadvantages of the client-server modelo Reliability

—The network depends on a possibly highly loaded server to function properly.

—Server needs to be replicated to some extent to provide better reliability.

o Scalability—More users imply more demand for computing

power, storage space and bandwidth

Peer-to-Peer Model

All nodes have same functional capabilities and responsibilitiesNo reliance on central services or resources.A node acts as both as a “server” and client.Considered more scalable

Peer-to-Peer Model

Peer-to-peer systems provide access to information resources located on computers throughout network.

Algorithms for the placement and subsequent retrieval of information is a key aspect of the design of P2P systems.

Why P2P?

The Internet has three valuable fundamental assetso Informationo Computing resources o Bandwidth

All of which are vastly under utilized,partly due to the traditional client-server model

Why P2P?

No single search engine can locate and catalog the ever-increasing amount of information on the Web in a timely way

Moreover, a huge amount of information is transient and not subject to capture by techniques such as Web crawlingo Google claims that it searches about 1.3x108

web pageso Finding useful information in real time is

increasingly difficult!

Why P2P?

Although miles of new fiber have been installed, the new bandwidth gets little use if everyone goes to Yahoo for content and to eBay

Instead, hot spots just get hotter while cold pipes remain cold

This is partly why most people still feel the congestion over the Internet while a single fiber’s bandwidth has increased by a factor of 10 6 since 1975, doubling every 16 months

Why P2P?

P2P potentially can eliminating the single-source bottleneck

P2P can be used to distribute data and control and load-balance requests across the Net

P2P potentially eliminates the risk of a single point of failure

P2P infrastructure allows direct access and shared space, and this can enable remote maintenance capability

Brief History

Generation 1 of P2P Systemso Napster music exchange

Generation 2o Freenet, Gnuetella, Kazaa, BitTorrent

Generation 3o Characterized by the emergence of

middleware layers for the application-independent management of distributed resources on a global scale

—Pastry, Tapestry, Chord, Kademlia

Environment Characteristics for Peer-to-Peer Systems

Unreliable environments Peers connecting/disconnecting – network

failures to participation Random Failures e.g. power outages,

cable and DSL failures, hackers Personal machines are much more

vulnerable than servers

Evaluating Peer-to-Peer Systems

A node’s database:o What does a node need to save in order to operate

properly/efficiently Success rate (if the file is in the network, what

are the changes that a search will find it) Lookup cost:

o Timeo Communication (bandwidth usage)

Join/departure cost Fault Tolerance – Resilience to faults Resilience to denial of service attacks, security.

Issues in File Sharing Services

Publish – How to insert a new file into the network

Lookup – Find a specific file Retrieval – Getting a copy of a file

P2P File Sharing Software

Allows a user to open up a directory in their file systemo Anyone can retrieve a file from directoryo Like a Web server

Allows the user to copy files from other users’ open directories:o Like a Web client

Allows users to search nodes for content based on keyword matches:o Like Google

Napster: How Did It Work

Application-level, client-server protocol over point-to-point TCP

Centralized directory server Steps:

o Connect to Napster servero Give server keywords to search the full list with.o Select “best” of correct answers.

—One approach is select based on the response time of a pings.

– Shortest response time is chosen.

Napster: How Did It Work

File list and IP address is uploaded

1.napster.com centralized directory

Napster: How Did It Work

napster.com centralized directory

Queryand

results

User requests search at server.

2.

Napster: How Did It Work

pingspings

User pings hosts that apparently have data.

Looks for best transfer rate.

3.napster.com centralized directory

Napster: How Did It Work

napster.com centralized directory

Retrievesfile

User choosesserver

4.

Napster’s centralized server farm had difficult time keeping up with traffic

Napster

There are centralized indexes but users supplied the files which were stored and accessed on their personal computer

Napster became very popular for music exchange

Napster

History: 5/99: Shawn Fanning (freshman,

Northeasten U.) founds Napster Online music service

12/99: first lawsuit 3/00: 25% UWisc traffic Napster 2/01: US Circuit Court of

Appeals: Napster knew users violating copyright laws

7/01: # simultaneous online users:Napster 160K, Gnutella: 40K,

Morpheus (KaZaA): 300K

Napster

Judge orders Napster to pull plug in July ‘01

Other file sharing apps take over!

gnutellanapsterfastrack (KaZaA)

8M

6M

4M

2M

0.0bit

s per

sec

Napster’s Downfall

Napster’s developers argued they were not liable for infringement of the copyrightso Why? They were not participating in the copying

process which was performed entirely between users’ machines.

This argument was not accepted by the courtso Why? The index servers were deemed an essential

part of the process Since the index servers were located at well-

known addresses, their operators were unable to remain anonymous.o Makes for an easy lawsuit target

Napster’s Downfall

A more fully distributed file sharing service spreads the responsibility across all of the users o Makes the pursuit of legal remedies difficult

Napster: Discussion

Locates files quickly Vulnerable to censorship and technical

failure Popular data become less accessible

because of the load of the requests on a central server

People started to look for more distributed solutions to file-sharing as a result of Napster’s failure.

Gnutella

Napster’s legal problems motivated Gnutella where there is not a use of centralized indexes

The focus is on a decentralized method of searching for fileso Central directory server no longer the bottlenecko More difficult to “pull plug”

Each application instance serves to:o Store selected fileso Route queries from and to its neighboring peerso Respond to queries if file stored locallyo Serve files

Gnutella

Gnutella history:o 3/14/00: release by AOL, almost immediately withdrawno Became open sourceo Many iterations to fix poor initial design (poor design

turned many people off) Issues:

o How much traffic does one query generate?o How many hosts can it support at once?o What is the latency associated with querying?o Is there a bottleneck?

Gnutella: Searching

Searching by flooding:• A Query packet might ask, "Do you have any content

that matches the string ‘Homer"? o If a node does not have the requested file, then 7

(default set by Gnutella) of its neighbors are queried. o If the neighbors do not have it, they contact 7 of

their neighbors.o Maximum hop count: 10 (this is called time-to-live

TTL)o Reverse path forwarding for responses (not files)

Gnutella: Searching

Downloading• Peers respond with a “QueryHit” (contains contact info)• File transfers use direct connection using HTTP

protocol’s GET method • When there is a firewall a "Push" packet is used –

reroutes via Push path

Gnutella: Searching

Gnutella: Searching

Gnutella: Discovering Peers

A peer has to know at least one other peer to send requests to.

Addresses of some peers have been published on a website.

When a peer enters the network, it contacts a designated peer and receives a list of other peers that have recently entered the network.

Gnutella: Discussion

Robust: The failure of peer is not a failure of Gnutella.

Performance: Flooding leads to poor performance

Free riders: Those who get data but do not share data.

The model of Gnutella just presented was found not be workable.

This led to models which had some peer nodes having indexes.

KaZaA: The Service

More than 3 million up peers sharing over 3,000 terabytes of content

More popular than Napster ever was More than 50% of Internet traffic ? MP3s & entire albums, videos, games Optional parallel downloading of files Automatically switches to new download server

when current server becomes unavailable Provides estimated download times

KaZaA: The Service

A user can configure the maximum number of simultaneous uploads and maximum number of simultaneous downloads

Queue management at server and cliento Frequent uploaders can get priority in server queue

Keyword searcho User can configure “up to x” responses to keywords

Responses to keyword queries come in waves; stops when x responses are found

KaZaA: The Technology

Proprietary Control data encrypted Everything in HTTP request and response

messages

KaZaA: Architecture

Each peer is either a supernode or is assigned to a supernode

o 56 min avg connecto Each SN has about 100-

150 childreno Roughly 30,000 SNs

Each supernode has TCP connections with 30-50 supernodes

o 23 min avg connect

supernodes

KaZaA: Architecture

Nodes that have more connection bandwidth and are more available are designated as supernodes

Each supernode acts as a mini-Napster hub, tracking the content and IP addresses of its descendants

A supernode tracks only the content of its children.

Considered a cross between Napster and Gnutella

KaZaA: Finding Supernodes

List of potential supernodes included within software download

New peer goes through list until it finds operational supernodeo Connects, obtains more up-to-date list, with 200 entrieso Nodes in list are “close” to ON.o Node then pings 5 nodes on list and connects with the

one If supernode goes down, node obtains updated

list and chooses new supernode

KaZaA Queries

Node first sends query to supernodeo Supernode responds with matcheso If x matches found, done.

Otherwise, supernode forwards query to subset of supernodeso If total of x matches found, done.

Otherwise, query further forwardedo Probably by original supernode rather than

recursively

Bootstrapping

How do I find out about a peer to begin with?

Use a bootstrapping (or multiple bootstrapping nodes).

Summary

The use of centralized indexes in Napster lands you in legal woes

The use of Gnutella avoids legal woes but is painfully slow.

Kazaa is somewhere in between. Can we do better?