DOC PREVIEW
U of I CS 414 - Media Server (Part 5)

This preview shows page 1-2-23-24 out of 24 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 24 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 24 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 24 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 24 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 24 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

CS 414 – Multimedia Systems Design Lecture 29 – Media Server (Part 5)AdministrativeOutlineTrue Video-On Demand SystemVOD System Delivery Schemes (to handle large number of clients)Caching for Streaming MediaTechniques for Increasing Server CapacityCachingCaching in Media ServersInterval CachingInterval CachingFrequency CachingTaxonomy of Cache Replacement PoliciesPatchingTypes of PatchingBatchingBatching PoliciesSlide 18Examples of Video Server SystemsSlide 20Factors affecting Optimal Block SizeSelecting MetricsSelecting MetricConclusionCS 414 - Spring 2011CS 414 – Multimedia Systems Design Lecture 29 – Media Server (Part 5) Klara NahrstedtSpring 2011AdministrativeMP3 is released (see class website for information)CS 414 - Spring 2011OutlineVOD Optimization Techniques Caching, Patching, BatchingExample of Multimedia File System - SymphonyTwo-level ArchitectureCello Scheduling Framework at Disk Management levelBuffer SubsystemVideo ModuleCachingCS 414 - Spring 2011True Video-On Demand SystemTrue VOD: serve thousands of clients simultaneously and allowing service any time (variable access time) Goal: minimize the required resource consumption such as Server bandwidth (disk I/O and network) – amount of data per time unit sent from server to clientsClient bandwidth – network bandwidth that a client must be able to receiveClient buffer requirements – amount of data client has to be able to temporarily store locallyStart-up delay – time between issuing request for playback and start of playback CS 414 - Spring 2011VOD System Delivery Schemes (to handle large number of clients)Periodic broadcast Data-centered approachServer channel is dedicated to video objects (movie channel) and broadcasting periodicallyScheduled multicastUser-centered approachServer dedicates channels to individual usersWhen server channel is free, the server selects a batch of clients to multicast according to some scheduling policyServer replication Servers maintaining the same videos are placed in multiple locations in the networkServer selection is a main issueCS 414 - Spring 2011Caching for Streaming MediaCaching – common technique to enhance scalability of general information disseminationExisting caching schemes are not designed for and do not take advantage of streaming characteristicsNeed New Caching for Streaming Media CS 414 - Spring 2011Techniques for Increasing Server Capacity CachingInterval CachingFrequency CachingKey Point In conventional systems, caching used to improve program performance In video servers, caching is used to increase server capacity CS 414 - Spring 2011Caching Read-ahead bufferingBlocks are read and buffered ahead of time they are neededEarly systems assumed separate buffers for each clientsRecent systems assume a global buffer cache, where cached data is shared among all clientsCS 414 - Spring 2011Caching in Media ServersCS 414 - Spring 2011Source: “preemptive, but safe interval caching for real-time multimedia systems “Lee et al. 2003Interval Caching This caching exploits sequential nature of multimedia accessesTwo streams Si and Sj are defined as consecutive if Si is the stream that next reads data blocks that have just been read by Sj. Such a pair of consecutive streams are referred to as preceding stream and following stream.Interval caching scheme exploits temporal locality accessing the same MM object, by caching intervals between successive streams (preceding stream and following stream)The interval caching policy orders all consecutive pairs in terms of increasing memory requirements. It then allocates memory to as many of consecutive pairs as possibleCS 414 - Spring 2011Interval CachingMemory requirements of intervals are proportional to length of interval and play-out rate of streams involvedWhen interval is cached, following stream does not have to go to disk, since all necessary data are in cache Algorithm:Order intervals based on increasing space –smaller interval implies smaller time to reaccessOptimal for homogeneous clientsDynamically adapts to changes in workloadCS 414 - Spring 2011Frequency Caching Typical video accesses follow 80-20 rule (i.e., 80% of requests access 20% of video objects)Cache most frequently accessed video objectsRequires large buffer spaceNot dynamicfrequency determination is based on past history or future estimates/Zipf distribution CS 414 - Spring 2011Taxonomy of Cache Replacement PoliciesRecency of access: locality of referenceFrequency based: hot sets with independent accessesOptimal: knowledge of the time of next accessSize-based: different size objectsMiss cost based: different times to fetch objectsResource-based: resource usage of different object classes CS 414 - Spring 2011PatchingStream tapping or patching – technique to support true VOD;Patching assumes multicast transmission and clients arriving late to miss the start of main transmissionThese late clients immediately receive main transmission and store it temporarily in a buffer. In parallel, each client connects to server via unicast and transports (patches) the missing video start which can be shown immediatelyCS 414 - Spring 2011Types of PatchingGreedy Patchinga new main transmission is started only if the old one has reached the end of the videoClients arriving in between create only patching transmissionsGrace PatchingIf a new client arrives and the ongoing main transmission is at least T seconds old, then the server automatically starts a new main transmission which plays whole video from the start again.CS 414 - Spring 2011Batching Batching – grouping clients requesting the same video object that arrives within a short duration of time or through adaptive piggy-backingIncreasing batching window increases the number of clients being served simultaneously, but also increases reneging probability reneging time – amount of time after which client leaves VOD service without delivery of videoIncreasing minimum wait time increases client renegingPerformance metrics: latency, reneging probability and fairnessPolicies: FCFS, MQL (Maximum Queue Length), FCFS-nCS 414 - Spring 2011Batching PoliciesFCFS: schedules the batch whose first client comes earliest, with the aim of achieving some level of fairnessMaximum


View Full Document

U of I CS 414 - Media Server (Part 5)

Documents in this Course
Lecture 1

Lecture 1

32 pages

LECTURE

LECTURE

30 pages

Load more
Download Media Server (Part 5)
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Media Server (Part 5) and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Media Server (Part 5) 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?