21
Layered Peer-to- Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaig n Source International Workshop on NOSSDAV’03, June, 2003

Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Embed Size (px)

Citation preview

Page 1: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Layered Peer-to-Peer Streaming

Yi Cui, Klara NahrstedtDepartment of Computer Science

University of Illinois at Urbana-ChampaignSource International Workshop on NOSSDAV’03, June, 2003

Page 2: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Outline

Problem addressingAsynchronousHeterogeneity

Layered peer-to-peer streaming solutionUnlimited number of supplying peersConstraint supplying peersLayered rate heterogeneity

Performance evaluation Conclusion

Page 3: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Problem addressing

AsynchronousUser request media data at different timeSolution: cache-and-relay approach

HeterogeneityRequest stream of different qualities due to

resource constraints such as network bandwidth

Solution: layer-encoded streaming approach

Page 4: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Problem addressing (cont.)

Page 5: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Layered peer-to-peer streaming

FeatureLimited inbound/outbound

bandwidth Goal

Maximize total qualities

Subject to

Qk: total receive layers of peer k

Ik: inbound bandwidth

Ok:outbound bandwidth

Hk:peers

Page 6: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Layered peer-to-peer streaming

lk: inbound bandwidth of Hk (# of layers) Ak: available layers at the cache of Hk

H0: server,

S = { H1, H2, …, HM } : set of hosts sorted by available layer number

i.e., A1≤ A2 ≤ … ≤ AM

Qkm: # of layers get from host m

Qk: streaming quality

Page 7: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Basic algorithm (cont.)

Page 8: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Basic algorithm

Available cache layers Outbound bandwidth

H1

H2

H3

H4

Hk

Get from server !

Allocate maximum # of layers for Hk

Page 9: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Enhanced algorithm

Ck: constraint on maximum # of supplying peers Qk*(M, Ck): optimal solution if Hk can only choose

Ck supplying peers from H1~HM

Qmax(Hm+1, …, HM):

best contributor in { Hm+1, …, HM } Maximize total receive data at Hk

DP implementation ─ O(CkM2)

Page 10: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop
Page 11: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Fault tolerance

Normal departureDue to user logout or quality degradationThe departure peer notifies Hk to reconfigure

FailDue to machine crash or network failureEither temporally request from server

or suffer from quality degradation

Page 12: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Fault tolerance (cont.)

Available cache layers Outbound bandwidth

H1

H2

H3

H4

Hk Access from server or degradation !

Page 13: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Layered rate heterogeneity

Layer rate allocation schemesNatural Number Scheme

l0 = r0, rk = k r‧ 0

Exponential Scheme rk = r0 2‧ k

Fibonacci Scheme r1 = 2r0, rk = rk-1 + rk-2

Page 14: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Layered rate heterogeneity (cont.) ri: streaming rate of layer I (Kbps)

Ik, Ok: inbound and outbound bandwidth (Kbps)

Page 15: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Performance evaluation

40000 peersModem/ISDN: 50%, 112Kbps (max) Cable Modem/DSL: 35%, 1Mbps (max)Ethernet peers: 15%, 10Mbps (max)

60-min video, which consists of 50 layers, with full quality streaming rate = 1Mbps

Run 24 hours

Page 16: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Overall streaming quality and scalability Streaming quality satisfaction = Qk/Ik

Page 17: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Tradeoff between overall quality & constrained supplying peers

Page 18: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Fairness

Page 19: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Robustness

Page 20: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Layer rate heterogeneity

Page 21: Layered Peer-to-Peer Streaming Yi Cui, Klara Nahrstedt Department of Computer Science University of Illinois at Urbana-Champaign Source International Workshop

Conclusion

Introduce a layered peer-to-peer streaming approach to optimize the streaming quality of heterogeneous peers, save server bandwidth.

Hope to make best use of bandwidth resource of supplying peers.

Evaluate the solution by: Test fairness among peers according to streaming

quality satisfaction and bandwidth contribution. Test robustness against unexpected departures/fails.