1. Peer to Peer Networking and Application
Hongfei Yan
Question: What is (Wireless / Computer) Networking?
Answer: In the world of computers, networking is the practice of linking two or more computing devices together for the purpose of sharing data. Networks are built with a mix of computer hardware and computer software.
Refer to Matei Ripeanu‘s slides

2. Today’s Objectives Understand real-world applications in terms of:
Motivation and objectives
Requirements: compute/storage/network resources
Architecture (“distributed systems” part)
Examples: Recent p2p applications
Start thinking of computer networks from the perspective of a networked-application
More intuitive

3. P2P Definition(s)
Def 1: “A class of applications that takes advantage of resources — storage, cycles, content, human presence — available at the edges of the Internet.”
Edges often turned on/off, without permanent IP addresses
Def 2: “A class of decentralized, self-organizing distributed systems, in which all or most communication is symmetric.”
Lots of other definitions that fit in between
Lots of (P2P?) systems that fit nowhere …
Def1:
Emphasis: on what resources are integrated
Problem: it is vague what one means by ‘edges’. A core network person would consider everything except routers and wires to be sitting on the edges of the network.
Example:
Def2:
Emphasis: how resources are integrated architectural / organizational solution chosen to integrate resources
Problem:
quite restrictive: (1) most people use P2P term for a lot of applications that do not fit this definition; (2) applications –like Gnutella- that were fitting this definition are moving away from it.
Vague again: “most communication is symmetric”
Example: Gnutella, DHTs (CAN, Chord, Tapestry, Pastry)

4. P2P Impact: Widespread adoption
Skype: >300M users
KaZaA – 389 millions downloads (1M/week) one of the most popular applications ever!
Number of users for file-sharing applications (estimate Sept ‘06)
eDonkey
3,108,909
FastTrack (Kazaa)
2,114,120
Gnutella
2,899,788
Cvernet
691,750 
Filetopia
3,405

5. P2P Impact (2): Huge resource users
P2P generated traffic now dominates the Internet load
Internet2 traffic statistics
Cornell.edu (March ’02): 60% P2P
UChicago estimate (March ‘01): Gnutella control traffic about 1% of all Internet traffic.

6. Floating Point Operations
P2P Impact (3) – Demonstrate that underused resources can be efficiently harnessed
Resources: CPU, storage space,
But also: network bandwidth, availability, user attention and expertise
statistics
Total
Last 24 Hours
Users
4,236,090
2,365
Results received
764M
1.13M
Total CPU time
1.3 M years
1.3 K years
Floating Point Operations
e+21
e+18 (51.40 TeraFLOPs/sec)

7. P2P Impact (4) – Social / Business
Data copying and distribution at (almost) zero almost cost
Might force a companies to change their business models
Digital content production and distribution
Telecomunications companies
But what business models?
P2P is great for ‘volunteer’ computing, yet its applicability is unclear for infrastructures

8. Roadmap Definitions Impact Applications Mechanisms
An In-depth Case Study

9. Applications: Number crunching
Examples: Entropia, UnitedDevices, etc
Characteristics (for
Massive parallelism
Low bandwidth/computation ratio
Fixed-rate data processing task
Error tolerance
Users do donate *real* resources
Problems
Centralized. Does it scale?
How to prevent cheating?
Approach suitable for a particular class of problems.
How to extend the model to problems that are not massively parallel
$1.5M / year
extra consumed power
$1.5M per year in consumed power

10. Applications: File sharing & content distribution
The ‘killer application’ to date
Too many to list them all:
BitTorrent, Napster, FastTrack (KaZaA, KazaaLite, iMesh), Gnutella (LimeWire, Morpheus, BearShare)
Two independent problems
Distributed index
Fast content download
Environment: unreliable, non-cooperative

11. Applications: Performance evaluation
Performance evaluation & monitoring requires multiple measurement points
Connectivity statistics
Routing errors
Evaluate Web-site performance form end-user perspective
Poor online performance costs businesses
$25 billion per year (Zone Research)
28% of attempted online purchases fail (BCG)
Slow page performance is the primary reason for transaction abandonment
Business transactions are at particular risk
Eight seconds was kind of the threshold, but that’s a ridiculous notion now.
The real expectations are around 4 seconds

12. Measurements: The Performance “Blind Spot”
Back-end Infrastructure
Network Landscape
Last-mile “Blind Spot”
Datacenter Testing
“Beacon”
Web server
ISP
Database
Backbone
Enterprise Provider
Firewall T1
Corporate
User
Corporate Network
ISP
App server
Backbone
3rd party content
Major Provider
Regional Network
Local ISP
Component Testing
Internet latency exists and for online businesses you must measure it.
Forrester, IDC, Gartner Research all say that many online customers will click out of your site or off of a page if that content does not down load within 5-8 seconds.
specific research bullet
Specific research bullet
50-80% of the internet latency impacting your customers occurs in the “last mile.” Research and IDC.
End to end web performance testing describes the measurement of your sites quality of service from your customers browser to your origin server.
Porivo’s distributed technology provides an active performance testing service that delivers true end to end web performance testing.
Datacenter
Monitoring
BMC
Mercury Interactive
Tivoli
ProactiveNet
HP OpenView
Computer Associates
Consumer User
Keynote Systems
Mercury Interactive
BMC/SiteAngel
Service Metrics
Critical to estimate end-to-end performance
Slide source:

13. Measurements: End-to-end Performance
Back-end Infrastructure
Network Landscape
Web server
Database 
ISP 
Backbone
Enterprise Provider
Firewall T1
Corporate
User
Corporate Network
ISP
App server
Backbone
3rd party content
Major Provider
Regional Network
Local ISP
Component Testing
Internet latency exists and for online businesses you must measure it.
Forrester, IDC, Gartner Research all say that many online customers will click out of your site or off of a page if that content does not down load within 5-8 seconds.
specific research bullet
Specific research bullet
50-80% of the internet latency impacting your customers occurs in the “last mile.” Research and IDC.
End to end web performance testing describes the measurement of your sites quality of service from your customers browser to your origin server.
Porivo’s distributed technology provides an active performance testing service that delivers true end to end web performance testing.
Datacenter
Monitoring
Consumer User
End-to-end Web Performance Testing
Slide source:
Slide source:

14. More applications … Backup storage (HiveNet, OceanStore)
Instant messaging (Yahoo, AOL)
Collaborative environments
Spam filtering
Anonymous
Censorship-resistant publishing systems (Ethernity, Freenet)

15. Roadmap Definitions Impact Uses and Examples Mechanisms
An In-depth Case Study

16. Mechanisms (I) To obtain a resilient system:
integrate multiple components with uncorrelated failure curves.
use replication for data and services to move the service ‘closer’ to the user

17. Example: Content Delivery Networks
Origin
Server ?
httpprx  ?
dnssrv
httpprx
Fetch data
from nearby
DNS Redirection
Return proxy,
preferably one
near client
Cooperative
Web Caching
Resolver
Browser
akamai.cnn.com

18. Mechanisms (II)
To improve the quality of the service delivered and reduce costs:
integrate multiple providers with uncorrelated demand curves
(lower over-provisioning at each of them)
move service delivery closer to the user

19. Example: Server consolidation
Averages utilization
40% Mainframes are idle
90% Unix servers are idle
95% PC servers are idle
But peak utilizations are …
0-15% Mainframes are idle in peak-hour
70% PC servers are idle in peak-hour
Source: “Grid Computing” Dr Daron G Green

20. Load Is Dynamic ibm.com external site February 2001
Daily fluctuations (3x)
Workday cycle
Weekends off
M T W Th F S S
World Cup soccer site
May-June 1998
Seasonal fluctuations
Event surges (11x)
Week

21. For Example: Energy-Conscious Provisioning
CPU idle
93w
CPU max
120w
boot
136w
disk spin
6-10w
off/hib
2-3w
work
watts
Idling consumes 60% to 70% of peak power demand.
Light load: concentrate traffic on a minimal set of servers
Step down surplus servers to low-power state
Activate surplus servers on demand
Even smarter: also manage air conditioning

22. EECE 411: Design of Distributed Software Applications
Mechanisms (III)
To provide anonymity:
use large number of independent components (“hide in the crowd”) and make search impossible (or at least costly)
Example: onion routing
EECE 411: Design of Distributed Software Applications

23. Mechanisms (IV) To detect anomalies, to generate good statistics:
Use multiple views
Example: Web-server performance characterization

24. Roadmap Definitions Impact Uses and Examples Mechanisms
An In-depth Case Study
File sharing – The Gnutella Network

25. Basic Primitives for File Sharing
Join: How do I begin participating?
Publish: How do I advertise my file(s)?
Search: How do I find a file?
Fetch: How do I retrieve a file?
Lots of different solutions for each of these four primitives. We’ll look at Gnutella network.

26. What makes Gnutella network interesting?
Large scale
today up to 2M nodes, 1000TB data, 100M files today
Self-organizing network
Fast growth in its early stages
more than 50 times during first half of 2001
(50 times again 2001 to 2006)
Open architecture, simple and flexible protocol
Interesting mix of social and technical issues
EECE 411: Design of Distributed Software Applications

27. Gnutella search mechanism
Boston
Chicago
MIT
UBC
Beatles: Yellow Submarine
Q:Beatles
Calgary
Gnutella nodes
TCP overlay tunnels
Routers
Search steps:
Initiates search for “Yellow Submarine”
Sends message to all neighbors
Neighbors forward message
Initiate reply message
Reply message is back-propagated
File download
I want to explain you briefly how Gnutella network works:
Gnutella nodes set up TCP tunnels to other existing Gnutella nodes. And all messages are forwarded on this overlay.
If a node at UBC is looking for a Beatles album … creates a query message and the query is “flooded” into the network.
We have build tools: to extract the topology of the Gnutella overlay, and to intercept the traffic.
EECE 411: Design of Distributed Software Applications

28. EECE 411: Design of Distributed Software Applications
Gnutella: Overview
Join: on startup, client contacts a few other nodes; these become its “neighbors”
Publish: no need
Search:
Flooding: ask neighbors, who is their neighbors, and so on... when/if found, reply to sender.
Back-propagation in case of success
Fetch: get the file directly from peer (HTTP)
[Note: this was the original design. Later the network moved to a two-layer structure]
EECE 411: Design of Distributed Software Applications

29. EECE 411: Design of Distributed Software Applications
Gnutella: Summary
Gnutella: self-organizing, large-scale, P2P application based on a hierarchical overlay network. It works!
Growth hindered by inefficient resource use.
I’ll summarize the this presentation:
Some solutions to help the network scale:
Organize the overlay network to match the underlying infrastructure topology.
Investigate methods for reducing traffic (query routing/filtering, better information organization).
Exploit locality in user interest  small world network (vorbit despre proiectul nostru de la Chicago)
Exploit caches  all while maintaining the self-organizing characteristics
EECE 411: Design of Distributed Software Applications

30. P2P: Summary Huge impact
Many architecture styles -- pros and cons for each
centralized, flooding, swarming,
Lessons learned:
Single points of failure are bad
Flooding messages to everyone does not scale
Underlying network topology is important
Not all nodes are equal
Need incentives to discourage freeloading
Privacy and security are important …

31. References
Eng Keong Lua, Jon Crowcroft, Marcelo Pias, Ravi Sharma and Steven Lim, "A survey and comparison of peer-to-peer overlay network schemes", IEEE Communications Surveys & Tutorials, (7)2: 22-73, Apr., 2005
An excellent survey of modern peer-to-peer systems, covering structured as well as unstructured networks.
This paper forms a good introduction for those wanting to get deeper into the subject but do not really know where to start.