2 Learning objectives To understand the nature of multimedia data and the scheduling and resource issues associated with it. To become familiar with the components and design of distributed multimedia applications. To understand the nature of quality of service and the system support that it requires. To explore the design of a state-of-the-art, scalable video file service; illustrating a radically novel design approach for quality of service. *
3 A distributed multimedia system Figure 15.1 Applications: –non-interactive: net radio and TV, video-on-demand, e-learning,... –interactive: voice &video conference, interactive TV, tele-medicine, multi-user games, live music,... *
4 Multimedia in a mobile environment Applications: –Emergency response systems, mobile commerce, phone service, entertainment, games,... *
5 Characteristics of multimedia applications Large quantities of continuous data Timely and smooth delivery is critical –deadlines –throughput and response time guarantees Interactive MM applications require low round-trip delays Need to co-exist with other applications –must not hog resources Reconfiguration is a common occurrence –varying resource requirements Resources required: –Processor cycles in workstations –and servers –Network bandwidth (+ latency) –Dedicated memory –Disk bandwidth (for stored media) At the right time and in the right quantities *
6 Application requirements Network phone and audio conferencing –relatively low bandwidth (~ 64 Kbits/sec), but delay times must be short ( < 250 ms round-trip) Video on demand services –High bandwidth (~ 10 Mbits/s), critical deadlines, latency not critical Simple video conference –Many high-bandwidth streams to each node (~1.5 Mbits/s each), high bandwidth, low latency ( < 100 ms round-trip), synchronised states. Music rehearsal and performance facility –high bandwidth (~1.4 Mbits/s), very low latency (< 100 ms round trip), highly synchronised media (sound and video < 50 ms). *
7 System support issues and requirements Scheduling and resource allocation in most current OS’s divides the resources equally amongst all comers (processes) –no limit on load – can’t guarantee throughput or response time MM and other time-critical applications require resource allocation and scheduling to meet deadlines –Quality of Service (QoS) management Admission control:controls demand QoS negotiation:enables applications to negotiate admission and reconfigurations Resource management: guarantees availability of resources for admitted applications –real-time processor and other resource scheduling *
8 The window of scarcity for computing and communication resources Figure 15.2 *
9 Characteristics of typical multimedia streams Data rate (approximate) Sample or frame frequency size Telephone speech64 kbps8 bits8000/sec CD-quality sound1.4 Mbps16 bits44,000/sec Standard TV video (uncompressed) 120 Mbpsup to 640x 480 pixelsx 16 bits 24/sec Standard TV video (MPEG-1 compressed) 1.5 Mbpsvariable24/sec HDTV video (uncompressed) 1000–3000 Mbpsup to 1920x 1080 pixelsx 24 bits 24–60/sec HDTV video MPEG-2 compressed) 10–30 Mbpsvariable24–60/sec Figure 15.3 *
10 Typical infrastructure components for multimedia applications : multimedia stream White boxes represent media processing components, many of which are implemented in software, including: codec: coding/decoding filter mixer: sound-mixing component * Figures 15.4 & 15.5 ComponentBandwidthLatencyLoss rateResources required Camera Out:10 frames/sec, raw video 640x480x16 bits Zero ACodecIn: Out: 10 frames/sec, raw video MPEG-1 stream InteractiveLow10 ms CPU each 100 ms; 10 Mbytes RAM BMixerIn: Out: 2 44 kbps audio 1 44 kbps audio InteractiveVery low1 ms CPU each 100 ms; 1 Mbytes RAM HWindow system In: Out: various 50 frame/sec framebuffer InteractiveLow5 ms CPU each 100 ms; 5 Mbytes RAM KNetwork connection In/Out:MPEG-1 stream, approx. 1.5 Mbps InteractiveLow1.5 Mbps, low-loss stream protocol LNetwork connection In/Out:Audio 44 kbpsInteractiveVery low44 kbps, very low-loss stream protocol This application involves multiple concurrent processes in the PCs Other applications may also be running concurrently on the same computers They all share processing and network resources
11 Quality of service management Allocate resources to application processes –according to their needs in order to achieve the desired quality of multimedia delivery Scheduling and resource allocation in most current OS’s divides the resources equally amongst all processes –no limit on load – can’t guarantee throughput or response time Elements of Quality of Service (QoS) management –Admission control:controls demand –QoS negotiation:enables applications to negotiate admission and reconfigurations –Resource management: guarantees availability of resources for admitted applications –real-time processor and other resource scheduling *
12 The QoS manager’s task * Figure 15.6 *
13 QoS Parameters Bandwidth –rate of flow of multimedia data Latency –time required for the end-to-end transmission of a single data element Jitter variation in latency :– dL/dt Loss rate –the proportion of data elements that can be dropped or delivered late * Protocol version Maximum transmission unit Token bucket rate Token bucket size Maximum transmission rate Minimum delay noticed Maximum delay variation Loss sensitivity Burst loss sensitivity Loss interval Quality of guarantee Bandwidth: Delay: Loss: Figure 15.8 The RFC 1363 Flow Spec acceptable jitter acceptable latency maximum rate burstiness percentage per T maximum consec- utive loss T value
14 Managing the flow of multimedia data Flows are variable –video compression methods such as MPEG (1-4) are based on similarities between consecutive frames –can produce large variations in data rate Burstiness –Linear bounded arrival process (LBAP) model: maximum flow per interval t = Rt + B(R = average rate, B = max. burst) –buffer requirements are determined by burstiness –Latency and jitter are affected (buffers introduce additional delays) Traffic shaping –method for scheduling the way a buffer is emptied * Protocol version Maximum transmission unit Token bucket rate Token bucket size Maximum transmission rate Minimum delay noticed Maximum delay variation Loss sensitivity Burst loss sensitivity Loss interval Quality of guarantee Bandwidth: Delay: Loss: Figure 15.8 The RFC 1363 Flow Spec acceptable jitter acceptable latency maximum rate burstiness percentage per T maximum consec- utive loss T value
15 (a) Leaky bucket * Figure 15.7 process 1 process 2 Traffic shaping algorithms – leaky bucket algorithm analogue of leaky bucket: –process 1 places data into a buffer in bursts –process 2 in scheduled to remove data regularly in smaller amounts –size of buffer, B determines: maximum permissible burst without loss maximum delay overkill?
16 Token generator (b) Token bucket * Figure 15.7 Traffic shaping algorithms – token bucket algorithm Implements LBAP –process 1 delivers data in bursts –process 2 generates tokens at a fixed rate –process 3 receives tokens and exploits them to deliver output as quickly as it gets data from process 1 Result: bursts in output can occur when some tokens have accumulated process 2 process 1 process 3 tokens: permits to place x bytes into output buffer
17 Admission control Admission control delivers a contract to the application guaranteeing: For each computer: cpu time, available at specific intervals memory Before admission, it must assess resource requirements and reserve them for the application –Flow specs provide some information for admission control, but not all - assessment procedures are needed –there is an optimisation problem: clients don't use all of the resources that they requested flow specs may permit a range of qualities –Admission controller must negotiate with applications to produce an acceptable result For each network connection: bandwidth latency For disks, etc.: bandwifth latency *
18 Resource management Scheduling of resources to meet the existing guarantees: Fair scheduling allows all processes some portion of the resources based on fairness: E.g. round-robin scheduling (equal turns), fair queuing (keep queue lengths equal) not appropriate for real-time MM because there are deadlines for the delivery of data Real-time scheduling traditionally used in special OS for system control applications - e.g. avionics. RT schedulers must ensure that tasks are completed by a scheduled time. Real-time MM requires real-time scheduling with very frequent deadlines. Suitable types of scheduling are: Earliest deadline first (EDF) Rate-monotonic e.g. for each computer: cpu time, available at specific intervals memory * EDF scheduling Each task specifies a deadline T and CPU seconds S to the scheduler for each work item (e.g. video frame). EDF scheduler schedules the task to run at least S seconds before T (and pre-empts it after S if it hasn't yielded). It has been shown that EDF will find a schedule that meets the deadlines, if one exists. (But for MM, S is likely to be a millisecond or so, and there is a danger that the scheduler may have to run so frequently that it hogs the cpu). Rate-monotonic scheduling assigns priorities to tasks according to tasks according to their rate of data throughput (or workload). Uses less CPU for scheduling decisions. Has been shown to work well where total workload is < 69% of CPU.
19 Scaling and filtering Source Targets High bandwidth Medium bandwidth Low bandwidth * Figure 15.9 Scaling reduces flow rate at source –temporal: skip frames or audio samples –spatial: reduce frame size or audio sample quality Filtering reduces flow at intermediate points –RSVP is a QoS negotiation protocol that negotiates the rate at each intermediate node, working from targets to the source.
20 QoS and the Internet Very little QoS in the Internet at present –New protocols to support QoS have been developed, but their implementation raises some difficult issues about the management of resources in the Internet. RSVP –Network resource reservation –Doesn’t ensure enforcement of reservations RTP –Real time data transmission over IP need to avoid adding undesirable complexity to the Internet IPv6 has some hooks for it IPv6 header layout *
21 Video on demand for a large number of users Quality of service Scalable and distributed Low cost hardware Fault tolerant Tiger design goals * Tiger Network Clients
22 Tiger architecture Storage organization –Striping –Mirroring Distributed schedule Tolerate failure of any single computer or disk Network support Other functions –pause, stop, start *
23 Tiger video file server hardware configuration * Each movie is stored in 0.5 MB blocks (~7000) across all disks in the order of the disk numbers, wrapping around after n+1 blocks. Block i is mirrored in smaller blocks on disks i+1 to i+d where d is the decluster factor n+10n+21n+32n+432n+1n Controller Cub 0Cub 1Cub 2Cub 3Cub n ATM switching network video distribution to clients Start/Stop requests from clients low-bandwidth network high-bandwidth Figure Cubs and controllers are standard PCs
24 Tiger schedule block play time T block service time t * Figure Cub algorithm: 1.Read the next block into buffer storage at the Cub. 2.Packetize the block and deliver it to the Cub’s ATM network controller with the address of the client computer. 3.Update viewer state in the schedule to show the new next block and play sequence number and pass the updated slot to the next Cub. 4.Clients buffer blocks and schedule their display on screen. slot 0 viewer 4 state viewer client viewer state: slot 1 free slot 2 free slot 3 viewer 0 state slot 4 viewer 3 state slot 5 viewer 2 state slot 6 free slot 7 viewer 1 state 012 in time t Stream capacity of a disk = T/t (typically ~ 5) Stream capacity of a cub with n disks = n x T/t Network address of client FileID for current movie Number of next block Viewer's next play slot Viewer state: slot 0 viewer 4 state viewer client viewer state: slot 1 free slot 2 free slot 3 viewer 0 state slot 4 viewer 3 state slot 5 viewer 2 state slot 6 free slot 7 viewer 1 state
25 Tiger performance and scalability 1994 measurements: –5 x cubs: 133 MHz Pentium Win NT, 3 x 2Gb disks each, ATM network. –supported streaming movies to 68 clients simultaneously without lost frames. –with one cub down, frame loss rate 0.02% 1997 measurements: –14 x cubs: 4 disks each, ATM network –supported streaming 2 Mbps movies to 602 clients simultaneously with loss rate of <.01% –with one cub failed, loss rate <.04% The designers suggested that Tiger could be scaled to 1000 cubs supporting 30,000 clients.
26 Summary MM applications and systems require new system mechanisms to handle large volumes of time-dependent data in real time (media streams). The most important mechanism is QoS management, which includes resource negotiation, admission control, resource reservation and resource management. Negotiation and admission control ensure that resources are not over-allocated, resource management ensures that admitted tasks receive the resources they were allocated. Tiger file server: case study in scalable design of a stream- oriented service with QoS.