1 USC INFORMATION SCIENCES INSTITUTE Proxy Caching Mechanism for Multimedia Playback Streams in the Internet R. Rejaie, M. Handley, H. Yu, D. Estrin USC/ISI WCW’99 April 1, 1999
2 USC INFORMATION SCIENCES INSTITUTE Motivation n Rapid growth in deployment of realtime streams(audio/video) over the Internet n Goals Š Maximize the quality of the delivered stream Š Minimize startup latency Š Low-latency VCR-functionality Š Minimize the load on the server & the network
3 USC INFORMATION SCIENCES INSTITUTE Outline n An End-to-end Architecture n Multimedia Proxy Caching n Conclusion n Future Directions
4 USC INFORMATION SCIENCES INSTITUTE Streaming Applications in Best-effort Networks (The Internet) n End-to-end congestion control is crucial for stability, fairness & high utilization Š Results in variable transmission rate n Streaming applications require constant average consumption rate Ô Streaming applications should be quality adaptive
5 USC INFORMATION SCIENCES INSTITUTE Quality Adaptation(QA) n Buffering only absorb short-term variations n Long-lived session could result in buffer overflow or underflow n QA is complementary for buffering Ô Adjust the quality(rate) with long-term variations Ô Layered framework BW(t) Time
6 USC INFORMATION SCIENCES INSTITUTE Buffer Manager Archive Error Control Quality Adaptation Transmission Buffer Cong. Control Acker Decoder Playback Buffer Internet ServerClient Adaptation Buffer Data path Control path The End-to-end Architecture Buffer Manager
7 USC INFORMATION SCIENCES INSTITUTE n Delivered quality is limited to the average bandwidth between the server and client n Solutions: Š Mirror servers Š Proxy caching Limitation Server Client Internet ISP Client Time L 0 L 1 L 2 L 3 L 4 Quality(layer)
8 USC INFORMATION SCIENCES INSTITUTE Server n Assumptions Š Proxy can perform: –End-to-end congestion ctrl –Quality Adaptation n Goals of proxy caching Š Improve delivered quality Š Low-latency VCR-functions Š Natural benefits of caching Proxy Internet Multimedia Proxy Caching Client
9 USC INFORMATION SCIENCES INSTITUTE Challenge n Cached streams have variable quality Ô Layered organization provides opportunity for adjusting quality Time L 0 L 1 L 2 L 3 L 4 Quality ( no. active layers ) Stored stream Played back stream
10 USC INFORMATION SCIENCES INSTITUTE Issues n Delivery procedure Š Relaying on a cache miss Š Pre-fetching on a cache hit n Replacement algorithm Š Determining popularity Š Replacement pattern
11 USC INFORMATION SCIENCES INSTITUTE Cache Miss Scenario n Stream is located at the original server n Playback from the server through the proxy n Proxy relays and caches the stream n No benefit in a miss scenario Server Internet Proxy Client
12 USC INFORMATION SCIENCES INSTITUTE Cache Hit Scenario n Playback from the proxy cache Š Lower latency Š May have better quality! n Available bandwidth allows: Š Lower quality playback Š Higher quality playback Server Proxy Internet Client
13 USC INFORMATION SCIENCES INSTITUTE Lower quality playback n Missing pieces of the active layers are pre- fetched on-demand n Required pieces are identified by QA n Results in smoothing Time L 0 L 1 L 2 L 3 L 4 Quality ( no. active layers ) Pre-fetched data Stored stream Played back stream
14 USC INFORMATION SCIENCES INSTITUTE n Pre-fetch higher layers on-demand n Pre-fetched data is always cached n Must pre-fetch a missing piece before its playback time n Tradeoff Time L 0 L 1 L 2 L 3 L 4 Quality ( no. active layers ) Pre-fetched data Stored stream Played back Stream Higher quality playback
15 USC INFORMATION SCIENCES INSTITUTE Replacement Algorithm n Goal: converge the cache state to optimal Š Average quality of a cached stream depends on –popularity –average bandwidth between proxy and recent interested clients Š Variation in quality inversely depends on –popularity Server Proxy Internet Client
16 USC INFORMATION SCIENCES INSTITUTE n Number of hits during an interval n User’s level of interest (including VCR- functions) n Potential value of a layer for quality adaptation Š Calculate whit on a per-layer basis n Layered encoding guarantees monotonically decrease in popularity of layers Popularity whit = PlaybackTime(sec) / StreamLength(sec)
17 USC INFORMATION SCIENCES INSTITUTE n Multi-valued replacement decision for multimedia object n Coarse-grain flushing Š on a per-layer basis n Fine-grain flushing Š on a per-segment basis Fine-grain Coarse-grain Cached segment Replacement Pattern Time Quality(Layer)
18 USC INFORMATION SCIENCES INSTITUTE Conclusion n End-to-end architecture for delivery of quality-adaptive multimedia streams Š Congestion control & Quality adaptation n Proxy caching mechanism for multimedia streams Š Pre-fetching Š Replacement algorithm Ô State of the cache converges to the optimal
19 USC INFORMATION SCIENCES INSTITUTE Future Directions n Extensive simulation(using VINT/ns) Š e.g. access pattern, the bandwidth distribution n Exploring other replacement patterns n Chunk-based popularity function
20 USC INFORMATION SCIENCES INSTITUTE Alternative Replacement Algorithm n Goal: to cache popular portion of each stream n Keep track of per-chunk popularity n Identify a victim chuck n Apply the same replacement pattern within the victim chunk