A Simple Model for Analyzing P2P Streaming Protocols

1
A Simple Model for Analyzing P2P Streaming
Protocols

• Zhou Yipeng Chiu DahMing John, C.S. Lui
• The Chinese University of Hong Kong

2
Outline

• Introduction
• Model Chunk Selection Strategies
• Simulation
• Conclusion

3
Introduction

• Unicast
• Client server is the bottleneck and waste
  bandwidth

Router
4
Introduction

• Application Layer Multicast (or CDN)
• Rely on a single distribution tree

Leaf peers
server
5
Introduction

• P2P Streaming System
• -P2P resolves this scalability problem by
  using all resources of all clients. It is like
  using multiple trees simultaneously to deliver
  content.

Peers maintain buffer neighbor list
6
Introduction

• P2P application
• -file distribution, p2p streaming
• Summary work on p2p streaming
• -PPlive, PPstream, CoolStreaming,
  BiTos
• -Much work on system study,
  architecture design and measurement but little
  theoretic work
• Our Contributions
• -Analytical Models on p2p streaming system
  to better understand
• -Chunk selection strategy study and a new
  strategy is proposed.
• -Trade off between continuity and
  scalability

7
Outline

• Introduction
• Model Chunk Selection Strategies
• Simulation
• Conclusion

8
Model Chunk Selection Strategies

• How buffer works?
• Server sends out chunks sequentially.
• Peer downloads one chunk every time slot
• Buffer shits ahead one position one time slot

server
playback
.
Buffer
9
Model Chunk Selection Strategies

• M peers with the same playback requirement
• Each has a playback buffer
• In each time slot, the server randomly selects
  one peer and uploads one chunk
• Users metric is the continuity, defined as p(n)
  , the probability chunk n available
• To compute p(n), recursively compute p(i). p(i)
  is defined as
• p(i)prob(position i filled)

10
Model Chunk Selection Strategies

• Each peers buffer is a sliding window
• In each time slot, each peer downloads a chunk
  from server or its neighbor
• q(i) the probability Bufi gets filled at this
  time slot, for igt1

p(1)1/M p(n)?
timet
sliding window
t1
p(1)1/M
11
Model Chunk Selection Strategies

• w(i) probability peer wants to fill Bufi
• w(i)1-p(i)
• h(i) probability the selected peer has the
  content for Bufi
• h(i)p(i)
• s(i) Bufi determined by chunk selection
  strategy

sliding window
p(1)1/M p(n)
1 2 i n
peer
neighbor
1 2 ... i n
p(1)1/M
12
Model Chunk Selection Strategies

• Greedy Strategy
• -try to fill the empty buffer closest to
  playback
• Rarest First Strategy
• -try to fill the empty buffer for the newest
  chunk since p(i) is an increasing function,
  this means Rarest First
• An example

13
Model Chunk Selection Strategies

• Greedy
• p(i1)p(i) (1-p(i)) p(i)
  (1-p(1)-p(n)p(i1))
• Rarest first
• p(i1)p(i) (1-p(i)) p(i) (1-p(i))
• Also studied
• continuous forms for these difference equations
  to study sensitivity
• Simulation to validate models

14
Model Chunk Selection Strategies

• From our models we can get the following
  conclusions
• Rarest First Strategy is more scalable than the
  Greedy Strategy as the peer population increases.
• The Greedy Strategy can achieve better continuity
  than Rarest First Strategy for small number of
  peers.

15
A New Chunk Selection Strategy

• Partition the buffer into 1,m and m1,n
• Use RF for 1,m first
• If no chunks available for download by RF, use
  Greedy for m1,n
• Difference equations become
• 
  
• 
  
  for i 1,,m-1
• 
  
• 
  
  for i m, n-1

16
Outline

• Introduction
• Model Chunk Selection Strategies
• Simulation
• Conclusion

17
Comparing Different Chunk Selection Strategies

• What do you mean by better?
• Playback continuity p(n) as large as possible
• Start-up Latency
• Given buffer size (n) and relatively large peer
  population (M)
• Rarest first is better in continuity!
• Greedy is the best in start-up latency
• Mixed is the best one of them

18
Simulation

• M1000
• N40
• In simulation,
• neighbors60
• Uploads at most 2 in each time slot for one peer
• Validate our model

19
Simulation
Rarest First
Mixed
Greedy

• 1000 peers, 40 buffer
• Compare three strategies, especially the curve
  for Mixed.

20
Simulation
RF
Mixed
Mixed
RF
Greedy
Greedy

• 1000 peers, buffer length varies from 20 to 40.
• For different buffer sizes
• Mixed achieves best continuity than both RF and
  Greedy
• Mixed has better start-up latency than RF

21
Simulation
RF
Greedy
RF
Greedy

• For (a), there are 40 peers. Greedy is better.
• For (b), the continuity requirement is fixed at
  0.93. RF is better

22
Simulation
Mixed

• Simulate 1000 peers, 2000 time slots
• Continuity is the average continuity of all peers
• Continuity for Mixed is more consistent, as well
  highest

23
Simulation
How to adapt m for the mixed strategy
Mixed
RF

• Adjust m so that p(m) achieves a target
  probability (e.g. 0.3)
• In simulation study, 100 new peers arrive every
  100 slots
• m adapts to a larger value as population increases

24
Outline

• Introduction
• Model Chunk Selection Strategies
• Simulation
• Conclusion

25
Conclusion

• Related work
• -Coolstreaming, BiTos
• Summary work on p2p streaming
• -There are many designed p2p streaming
  systems, such as PPLive, PPstream
• -Many measurement papers on these system
• -Little work on model analysis
• -Little study on chunk selection strategies
• Our Contribution
• -Analytical Models on p2p streaming system
  to better understand
• -Chunk selection strategy study
• -Mixed strategy is proposed, which is better
  than RF or Greedy
• -Trade off between continuity and
  scalability

26
Thank You