An adaptive video multicast An adaptive video multicast scheme for varying workloadsscheme for varying workloads
Kien A.Hua, JungHwan Oh, KhanKien A.Hua, JungHwan Oh, Khanh Vuh Vu
Multimedia Systems, Springer-Verlag 2002Multimedia Systems, Springer-Verlag 2002
OutlineOutline
• Introduction
• Related Work
• Proposed Approach
• Performance Model
• Performance
• Conclusion
IntroductionIntroduction
• Maximize the efficiency of server resource with– Periodic broadcast – Scheduled Multicast (batching)– Hybrid Design
• Will show that existing scheduled multicast techniques are not suited for hybrid designs
Related WorkRelated WorkSkyscraper Broadcasting SchemeSkyscraper Broadcasting Scheme
• Fragmentation recursive function
– Series [1,2,2,5,5,12,12,25,25,52,52,…]
Related WorkRelated WorkSkyscraper Broadcasting SchemeSkyscraper Broadcasting Scheme
Apply aforementioned series function => segment size = 10.4 !!
22-mins video
1.5 Mbits/sec
5 min buffer
0.5 min latency
Related WorkRelated Work Scheduled Multicast Scheduled Multicast
• Differ primarily in the criterion used to select which batch will receive service– First come, first served (FCFS)– Maximum queue length first (MQL)– Maximum factored queue length first (MFQ)
Related WorkRelated Work Scheduled Multicast Scheduled Multicast
• FCFS
• MQL
timeVideo 1
timeVideo 2
time
timeVideo 2
Video 1
Fair , maybe bad throughput
Better throughput, maybe not fair
Related WorkRelated Work Maximum factored queue length first (MFQ)Maximum factored queue length first (MFQ)
• Applying a discriminatory weighting factor to the length of the queue
• : video i length, : the request frequency of video i
• Schedule video with the largest value of
iq if
tqtq
q
f
qi
i
i
i
i
/
Related WorkRelated Work Maximum factored queue length first (MFQ)Maximum factored queue length first (MFQ)
• d• d
Still not fair, because not average waiting time !!
Proposed ApproachProposed Approach
• Adaptive Hybrid Approach (AHA)– With a novel scheduled multicast --“Largest aggregated waiting time first scheme” (LAW)
– And SB (skyscraper broadcast)
LAWLAW
• Compare with MQL, it take account of the distribution of the request – With considering “aggregated waiting time”
m : the total number of pending requests for video i
ija : the arrival time of the jth request for video i
c : the current time
LAWLAW
S1=128*5-(107+111+115+121+126)=60
S2=128*4-(112+119+122+127)=32
S1=128*5-(107+111+115+121+126)=60
Compute the sum of video i service latency
Adaptive hybrid approachAdaptive hybrid approach
• With following procedures to decide which videos to broadcast
Adaptive hybrid approachAdaptive hybrid approach
Performance modelPerformance model
• Compare AHA with MFQ-SB-n
• Performance metrics– Defection rate– Unfairness– Average service latency– Throughput
• 100 videos, each 120 mins,
avg. playback rate 1.5 Mbits/sec.
Performance modelPerformance model
Performance - LAW vs. MFQPerformance - LAW vs. MFQ
Arrival rate : 8 req/min
Skew factor : 0.3
LAW perform slightly better than MFQ
in service latency,throughput, defection rate
PerformancePerformance
PerformancePerformance
• Compare MFQ-SB-n with altering one of– Server Capacity (channels)– Request Arrival Rate– Skew Factor
Alter Server Capacity Alter Server Capacity (channels)(channels)
Alter Request Arrival RateAlter Request Arrival Rate
Alter Skew FactorAlter Skew Factor
Alter request rate & skew factorAlter request rate & skew factor
ConclusionConclusion
• Prove that existing scheduled multicast schemes are not suited for hybrid design
• Proposed a new technique called Largest Aggregated Waiting time first (LAW)
• AHA is capable of coping with a changing workload
Periodic Broadcast (1996)Periodic Broadcast (1996)
video
…
…
…
time
• PB v.s. batch:– Short initial delay– Large client-side buffer
Client requests
Batching (1993)Batching (1993)
0 t1 t2 t3
Client requests
time
• Batch window:– The time interval to initiate a batch stream.
Adaptive hybrid approachAdaptive hybrid approach
Adaptive hybrid approachAdaptive hybrid approach