Download Problem Statement

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Wireless security wikipedia , lookup

IEEE 802.1aq wikipedia , lookup

Video on demand wikipedia , lookup

Extensible Authentication Protocol wikipedia , lookup

Piggybacking (Internet access) wikipedia , lookup

Computer network wikipedia , lookup

AppleTalk wikipedia , lookup

SIP extensions for the IP Multimedia Subsystem wikipedia , lookup

Airborne Networking wikipedia , lookup

Net bias wikipedia , lookup

Cracking of wireless networks wikipedia , lookup

Recursive InterNetwork Architecture (RINA) wikipedia , lookup

List of wireless community networks by region wikipedia , lookup

Zero-configuration networking wikipedia , lookup

Internet protocol suite wikipedia , lookup

Deep packet inspection wikipedia , lookup

Hypertext Transfer Protocol wikipedia , lookup

Streaming media wikipedia , lookup

Peer-to-peer wikipedia , lookup

Transcript
Problem Statement of P2P Streaming
Protocol (PPSP)
draft-ietf-ppsp-problem-statement-01
Y. Zhang, N. Zong, G.Camarillo, J.seng and R. Yang
IETF-80, Prague, March 28 , 2011
1
Problems
• Hard to share resource with multiple private
protocols
– Memory
– Storage
– Bandwidth
ISPs, vendors and terminals
• Hard to adapt with new environment current
protocols don’t address
– Including mobile and wireless network with a
different characteristic in both network and
terminals
2
What’s new in the PS(1)
• Difficulties in building open streaming delivery infrastructure
with lots of private protocols
• ISP has the willing to build an open infrastructure for low-cost
unified streaming delivery using P2P tech (ISP owned P2Ped CDN)
• Also current CDN using protocol like HTTP is costly for streaming vendors
• But private P2P streaming protocols lead to
• Vendor deploys its own P2Ped CDN network
• Storage and traffic waste in the ISP for same content as a whole
• Worse when P2P streaming traffic percentage is increasingly higher
A
B
X
…
Storage: X times
ISP
Traffic in backbone:X times
A’s own B’s
… own X’s own
CDN
CDN CDN
User
3
What’s new in the PS(2)
• Terminal physical resource starvation with lots of
private protocols
– iPAD: 256M memory, 16G storage
– iPhone(X generation): 20M available memory in practice
– Current P2P Streaming occupation:~100M memory and
~1G storage
100M, 1G
200M, 1G
– Concurrent running scenarios
…
• PPStream for live streaming and PPVA
for helping others (only contributing)
4
What’s new in the PS(3)
•Difficulties in mobile environment for using
current protocols
–Any difficulties?
•Performance degradation
•Adaptation: what kinds of mobile terminal and network
information to carry in tracker and peer protocol for better
performance
•Terminal capability
•Network dynamics
−Question in ML: Is mobile network so broad to accommodate P2P streaming?
−3G:Already 30% traffic are P2P in some networks
−LTE: uplink:50Mbps downlink:100Mbps
5
Open tracker and peer protocol enable
memory, storage and bandwidth sharing and
saving for same content in both terminal and
network sides with reduced infrastructure
deployment cost among different streaming
applications
Open tracker and peer protocol addresses
fixed and mobile/wireless converged network
environment
6
Use case updates(1-a)
• Interworking between different streaming providers
– Currently: IP outside certain region cannot access some
P2P streaming (by policy) or has a bad performance
• Limit or wrong knowledge on out of scope IP addresses
– Cooperation can solve this problem
– Loose coupling: Tracker and normal peer, with different
software and scheduler
Tracker
protocol
Tracker
protocol
Peerlist request/response
SPA
SPA
SPB
Request miss redirecting SPB
Request hit:
Subscribe B
SubscriberA
SPB
Request miss Peerlist in SPB
schedulerA
SubscriberB
Peer protocol
SubscriberA
schedulerB
SubscriberB
7
Use case updates(1-b)
• Tight coupling: Tracker and SuperNodes sync
SNA
Tracker A
SPA
SPB
Req/res SNB
Request miss
SNB
Tracker B
Request hit:
SNA
SubscriberA
8
Use case updates(2-a)
• Open ISP’s CDN supporting P2P streaming
with tracker protocol
– Edge nodes deployment saving: Some ISPs attract
SPs with very cheap/ even free speeding
Avada.flv
SPA
Hero.mp4
SPB
Hereafter.avi
SPX
For same content
Deployed Storage: 1 times
Traffic in backbone:1 times
Tracker
protocol
ISP
Largely reduce the storage
And traffic waste
User
9
Use case updates(2-b)
• Hybrid CDN-P2P delivery with peer protocol
– Tight coupling (Newly built CDN):
• No difference between CDN nodes and peers
• Trackers can act as the scheduler in the CDN
• Building CDN network is just the same as building p2p overlay
– Loose coupling( Existing CDN):
• Dual stack for http and ppsp in CDN nodes
• Easily separate the distribution (http based) and delivery (ppsp
tracker and peer protocol based)
CDN
Peer protocol
Tight couple
User
HTTP protocol
CDN
Peer protocol
User
10
Loose couple
Use case updates(3)
•Single client-base supporting different apps
Specified Codec
Specified encryption
User2 PPSP compatible client
for App1
User3 PPSP compatible client
for App2
App1
App2
Peer protocol
Peer protocol
Tracker protocol
Scheduling Scheduling
2
1
Shared Memory
Shared Storage
User1 PPSP compatible client
(Dameon)
Use the tracker and peer protocol for necessary information in streaming
acquisition and sharing
11
Use case updates(4)
• Open Video Acceleration (VA) with converged http streaming
and PPSP streaming
Web Site1
Web Site N
HTTP request
for A.flv
HTTP
agent
HTTP stream
A.flv
P2P VA
Plug-in
Tracker2
A.flv
Peerlist:node2,node3,..
P2P VA
Plug-in
Node1
Node 2
P2P stream
P2P VA
Plug-in
Node 3
12
Security part update
• Adding the consideration on untrusted peers
– PPSP security considerations involve the security problems
related to the introduction of p2p technology (e.g. usage
of untrusted peers) and the used PPSP protocols.
• Malicious peers DDoS attack to tracker by sending fake request
• Malicious peers may report fake information (e.g.,cheating
trackers and other peers by claiming itself owning some unexisting
data).
• User authentication and data integrity check for
streaming may be necessary for PPSP
• Do we need a draft on this?
13
Next step
• Modify according to the suggestions and
comments
• Ask for WGLC
14
Thanks for your attention!
Q&A?
15
Motivation
• What does integrity mean in P2P streaming?
– The media content is exactly the same as published from
a certain source and not manipulated by any
intermediate party in the network.
• Why do we need to protect media content’s
integrity?
– Desirable from the media publisher’s point of view
• Who holds certain reputation/authority/responsibility for the
media content's authenticity/validity it provides to the public.
– Desirable from the downloading peer’s point of view
• To ensure the received media is authentic from a valid source.
16
Proposal
17
Open issues
• Which type of certificate should be used?
– Certificate for the publishing entity, peer, or
program?
• Who should be responsible for the certificate
distribution?
– The tracker or the peers?
• Who should issue the certificate?
– Publishing entity, peer, tracker or a trusted third
party?
18