Transcript PPT - Mobile Multimedia Laboratory

Realistic Media Streaming over BitTorrent
George Xylomenos
Mobile Multimedia Laboratory
Greece
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Outline
•
•
•
•
•
•
•
•
•
Context
Motivation
Advantages
Streaming over BitTorrent
Experimental setup
Number of stall periods
Average stall duration
Download and seeding time
Conclusion
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Context
• The ICT PURSUIT Project
– The Internet mostly disseminates data
– Publish-Subscribe Internet Technology
– Clean slate approach to Future Internet
• What does that have to do with BitTorrent?
– PURSUIT was motivated by content distribution
– BitTorrent is the perfect benchmark for this!
– Now if we only had a good BitTorrent simulator…
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Motivation
• Why a BitTorrent simulator?
– BitTorrent swarms exhibit very complex behavior
• Many mechanisms and strategies are at play
– Traces are hard to gather and understand
• Fully distributed systems are hard to monitor
– Performance predictions are simply guesses
• What happens when we modify a strategy?
– Existing simulators were not detailed enough
• Some omit large parts of the protocol
• Others only work over abstract networks
• Most are custom-built and hard to extend
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Advantages
• Why bother with our simulator?
– It operates over the OMNeT++ platform
• You can use it with everything available for OMNeT++
• Example: OMNeT++ supports two types of network
– InetUnderlay: hosts with full TCP/IP stacks
– SimpleUnderlay: simple and fast abstract hosts
– It incorporates nearly all BitTorrent details
• All policies and options are present and tunable
• If something is missing, feel free to add it!
– Extra features that simplify simulations
• Asymmetric links, churn model, GT-ITM topologies
– To probe further
– K. Katsaros, V. Kemerlis, C. Stais, and G. Xylomenos, A BitTorrent
module for the OMNeT++ simulator, IEEE MASCOTS, 2009
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Streaming over BitTorrent
• A case in point: Streaming over BitTorrent
– A download window favors “sequential” downloads
– Window moves on playback or piece download
• If a piece is not available for playback, the player stalls
– Three different proposals exist, but…
• Undocumented simulation setups
• Different evaluation metrics
• Retransmissions and congestion are ignored
– This paper avoids all these problems
• Realistic simulations with the same assumptions
– To probe further
– C. Stais, G. Xylomenos, and A. Archontovasilis, A comparison of
streaming extensions to BitTorrent, IEEE ISCC, 2011
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Streaming over BitTorrent
• Fixed-Size Window (FSW)
– Fixed sliding window from first non-available piece
– Rarest-first only within the window
• High-Priority Set (HPS)
– Fixed size window of non-downloaded pieces
– Download outside the window with probability 1-p
• Stretching Window (SW)
– Also uses an HPS but only downloads inside it
– Bounded distance between first and last piece
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Streaming over BitTorrent
Player
Begin
OK
OK
D
1
2
3
End
OK
OK
5
6
4
7
8
FSW
Player
Begin
OK
OK
D
1
2
3
HPS
4
3478
End
OK
OK
5
6
7
8
HPS
Player
Begin
OK
OK
D
1
2
3
HPS
4
347
End
OK
OK
5
6
7
8
SW
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Experimental setup
• Main parameters
– Video: 200 MB @ 256 kbps (106 min)
• 112 KB pieces (1 GOP = 3.5 seconds)
– 4 core / 192 access routers
– ADSL access links: Uplink 1-2 Mbps, Download 4-24 Mbps
– 1 seeder and 120 peers, incremental joins
• 50% of peers stay in the swarm until playback completes
– Prefetch 1 or 5 pieces before playback
– Window size 2% or 8% of file (36 or 147 pieces)
– HPS probability 80%
– SW limit 50 or 200 pieces
– Two player modes
• B1: the player stalls until the next piece completes
• B2: the player stalls until the next three pieces complete
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Number of stall periods
4.50
3.50
HPS: 1p & 2%
4.00
HPS: 5p & 2%
3.50
HPS: 1p & 2%
3.00
HPS: 5p & 2%
HPS: 1p & 8%
HPS: 5p & 8%
3.00
HPS: 1p & 8%
2.50
HPS: 5p & 8%
FSW: 1p & 2%
2.50
FSW: 5p & 2%
2.00
FSW: 1p & 8%
FSW: 1p & 2%
2.00
FSW: 5p & 2%
FSW: 1p & 8%
1.50
FSW: 5p & 8%
1.50
SW: 1p & 2%
1.00
SW: 5p & 2%
0.50
SW: 1p & 8%
FSW: 5p & 8%
SW: 1p & 2%
1.00
SW: 5p & 2%
0.50
SW: 1p & 8%
SW: 5p & 8%
0.00
SW: 5p & 8%
0.00
• How many times will a user experience a stall?
– 30% more when we stall only for a single piece (left)
– In all cases, very few stalls for 100+ minutes
– HPS is the worst performer, despite its complexity
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Average stall duration
300.00
400.00
HPS: 1p & 2%
HPS: 5p & 2%
250.00
HPS: 1p & 8%
HPS: 1p & 2%
350.00
HPS: 5p & 2%
300.00
HPS: 1p & 8%
HPS: 5p & 8%
200.00
FSW: 1p & 2%
FSW: 5p & 2%
150.00
FSW: 1p & 8%
FSW: 5p & 8%
100.00
HPS: 5p & 8%
250.00
FSW: 1p & 2%
FSW: 5p & 2%
200.00
FSW: 1p & 8%
150.00
FSW: 5p & 8%
SW: 1p & 2%
SW: 5p & 2%
50.00
SW: 1p & 8%
SW: 1p & 2%
100.00
SW: 5p & 2%
SW: 1p & 8%
50.00
SW: 5p & 8%
0.00
SW: 5p & 8%
0.00
• How long does each stall last?
– 30% more when we stall for three pieces (right)
– Stall durations vary from noticeable to quite long
– HPS is the worst by far with a larger window
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Download and seeding time
4000.00
3000.00
HPS: 1p & 2%
3500.00
HPS: 5p & 2%
3000.00
HPS: 1p & 8%
HPS: 5p & 8%
2500.00
HPS: 1p & 2%
HPS: 5p & 2%
2500.00
HPS: 1p & 8%
HPS: 5p & 8%
2000.00
FSW: 1p & 2%
FSW: 5p & 2%
2000.00
FSW: 1p & 2%
FSW: 5p & 2%
1500.00
FSW: 1p & 8%
1500.00
FSW: 5p & 8%
SW: 1p & 2%
1000.00
FSW: 1p & 8%
FSW: 5p & 8%
1000.00
SW: 1p & 2%
SW: 5p & 2%
SW: 1p & 8%
500.00
SW: 5p & 2%
500.00
SW: 1p & 8%
SW: 5p & 8%
0.00
SW: 5p & 8%
0.00
• Download time: same for B1 and B2
– Better with HPS, but always lower than the video duration
• Seeding time: same for B1 and B2
– Also better with HPS, as the download finishes earlier
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory
Conclusion
• Which scheme is best?
– HPS leads to visibly worse user experience
• More stalls with longer durations
– FSW and SW are acceptable and very close
• SW is not worth the extra complexity over FSW
– FSW leads to 1-2.5 stalls of 40-60 seconds for 100+ minutes
– The two buffering modes offer the expected tradeoff
• Either fewer or longer stall periods
• Future work
– Explore additional parameters and tricks
– Study individual peer performance
Session 8b, 5th July 2012
Future Network & MobileSummit 2012
Copyright 2012 Mobile Multimedia Laboratory