Proxy-Assisted Techniques for Delivering Continuous Multimedia StreamsAgendaRelated WorkLimitations of current technologyProxy-Assisted Video Delivery ArchitectureAdvantages of proxy-assisted video deliveryClassificationAdvantages of proposed architecturesProxy-Assisted CatchingSlide 10OptimizingTerms involvedCalculationCalculation contd..Controlled MulticastComparison with Proxy-Assisted Controlled MulticastObservationProxy-Assisted Selective CatchingClassifying “Hot” and “Cold” videosSimulation resultsAssumptionsWaiting time vs. total number of channelsWaiting time vs. Arrival rateTotal no. of channels vs. arrival rateWaiting time vs. Server channelsNumber of channels vs. Arrival rateSlide 27ConclusionSlide 291Proxy-Assisted Techniques for Delivering Continuous Multimedia StreamsLixin Gao, Zhi-Li Zhang, and Don Towsley2AgendaRelated workProxy-Assisted Video Delivery ArchitectureProxy-Assisted CatchingProxy-Assisted Selective CatchingSimulation resultsConclusion3Related WorkMulticast TechniquesClient pull Server-pushBatchingPatchingServer-push -> Typically designed for “hot” (frequently requested) objects -> Fixed number of multicast channels4Limitations of current technologyServer and network resources (Server I/O bandwidth and network bandwidth) are major limiting factors in widespread usage of video streaming over the internetNeed techniques to efficiently utilize server and network resourcesService latency and popularity of video object should be considered5Proxy-Assisted Video Delivery Architecture6Advantages of proxy-assisted video deliveryLatency reduction without increasing demand on backbone network resourcesNeed to store only the initial frames hence feasible with large data volumeI/O bandwidth requirement on proxy server is insignificant, since responsible for limited number of clients7ClassificationProxy-assisted video delivery architectureProxy-assisted catchingProxy-assisted Selective catchingProxy-assisted catching : Suited for “hot” video objectsProxy-assisted selective catching : Even suited for “cold” (less frequently requested) video objects8Advantages of proposed architectures Reduce the resources requirements at central serverReduce service latency experienced by clients Assumptions Client can receive data from 2 channels simultaneously9Proxy-Assisted CatchingReduces service latency by allowing clients to join an ongoing broadcastClients catch-up by retrieving initial frames using unicast channel from proxy10Proxy-Assisted CatchingPartition function used11Optimizing Server and network bandwidth are major bottleneck. Hence reducing total number of channels requiredTrade-off between -> Number of dedicated channels by server -> Storage space required by proxy12Terms involvedN : No. of video objects on central serverL : Length of video λ : Request rate (Poisson distribution)K : Server channels to broadcast videoK* : Optimal number of server channelsi : Video object no.j : Broadcasting frame13Calculation No. of proxy channels required : Total no. of channels required : Tradeoff between number of server channels and expected number of proxy channels required for catch-up14Calculation contd..Optimization problem : Expected number of channels : Optimal no. of server channelsOptimal no. of proxy channels15Controlled MulticastClient pull techniqueAllows client to join the ongoing multicast if it requests with a certain threshold time TiElse a new multicast channel is allocatedProxy-assisted Controlled MulticastProxy pre-store the initial Ti frames of videoMissing portion of video is send separately through a unicast channelGood technique for “cold” video objects16Comparison with Proxy-Assisted Controlled MulticastTotal no. of channels required for controlled multicast is : For large value of λ no. of channels required by proxy-assisted catching is lessVerified using following setup : L : 90 min. video object17Observation0.418Proxy-Assisted Selective CatchingCombines Proxy-Assisted Catching and Controlled MulticastBroadcast most frequent videos using Proxy-Assisted Catching and less frequent videos using Controlled Multicast19Classifying “Hot” and “Cold” videosHot video ifTotal no. of channels required using catchingTotal no. of channels required using controlled multicast20Simulation resultsSimulation settings N : No. of video objects on central server λ : Request rate (Poisson's distribution)Simulates 150 hours of client requestsKi* : Broadcasting channels for “hot” video objectsRemaining channels for controlled multicastFirst-come-first-serve basis21AssumptionsSufficient proxy resources to store prefixes for all videosProxy server has 40GB of storage space and I/O bandwidth of 88 Mb/s22Waiting time vs. total number of channels710900λ = 5023Waiting time vs. Arrival rateλ varies from 40 to 80Total no. of channels = 70024Total no. of channels vs. arrival rate100150Performance of selective catching and catching same25Waiting time vs. Server channels46070036% saving in number of channels required at central server26Number of channels vs. Arrival rateSignificant reduction in central server channel requirement27Waiting time vs. Server channelsAdvantage of proxy-assisted selective catching does not critically depend on availability of proxy storage space28ConclusionApproach is proved using quite realistic simulations without any major assumptionsIf the arrival rate exceeds beyond certain assumptions then the service latency will
View Full Document