Resilient Peer-to-Peer Streaming Presented by: Yun Teng

Resilient Peer-to-Peer Streaming Authors Venkata N. Padmanabhan Venkata N. Padmanabhan Helen J. Wang Helen J. Wang Philip A. Chou Philip A. ChouFrom Microsoft Microsoft

Motivation  Distributing “live” streaming media content to a potentially large and highly dynamic population of hosts. “Live” streaming refer to the simultaneous distribution of the same content to all clients “Live” streaming refer to the simultaneous distribution of the same content to all clients

Challenge  Peer-to-peer content distribution is attractive because the bandwidth available to serve content scales with demand.  A key challenge: making content distribution robust to peer transience.

Approach  Introduce redundancy, both in network paths and in data  Multiple diverse distribution tree: provide redundancy in network paths  Multiple description coding (MDC): provide redundancy in data

CoopNet  Makes selective use of P2P networking, placing minimal demands on the peers  Goal: help a server tide over crises such as flash crowds rather than replace the server with a pure P2P system

Assumption  A node participates in and contributes bandwidth for content distribution only so long as the user is interested in the content. It stops forwarding traffic when the user tunes out  A node only contribute as much upstream bandwidth as it consumes in the downstream direction (applies to the total bandwidth in and out of a node aggregated over all trees  Nodes in CoopNet are inherently unreliable

Tree management

Goals  Short trees  Tree diversity  Efficiency  Quick join and leave  Scalability

Conflicts  Tree diversity versus efficiency  Quick join and leave versus scalability

Feasibility of the Centralized Protocol  September 11 flash crowd at MSNBC At peak, 18,000 nodes, 1,000 arrivals and departures per second At peak, 18,000 nodes, 1,000 arrivals and departures per second On average, 10,000 nodes, 180 arrivals and departures per second On average, 10,000 nodes, 180 arrivals and departures per second  Resource requirement Memory: 10 MB Memory: 10 MB Network bandwidth: 8 Mbps Network bandwidth: 8 Mbps CPU: 40 ns memory cycle, allow 390 memory accesses per insertion CPU: 40 ns memory cycle, allow 390 memory accesses per insertion

Centralized Tree Management  Randomized Tree Construction  Deterministic Tree Construction

Tree Efficiency / Topology Awareness  Need an efficient way to pick a proximate parent for a node without requiring extensive P2P network measurements  Each node maintains its “delay coordinates” of ping times to a small set of landmark hosts  Root pick the closest node for incoming node from a set of candidate parents

Tree Repair  Due to node leave  Two types Voluntary – Notify the root Voluntary – Notify the root Failure – Detect failure Failure – Detect failure

Multiple Description Coding (MDC)

Overview  Encoding an audio and/or video signal into M>1 separate streams, or descriptions, such that any subset of these descriptions can be received and decoded.  The distortion with respect to the original signal is commensurate with the number of descriptions received.

Overview (cont.)  MDC incurs a modest performance penalty relative to layered coding, which in turn incurs a slight performance penalty relative to single description coding.  The audio and/or video signal is partitioned into groups of frames (GOF), each group having duration of T (such as 1 second). Each GOF is independently encoded, error protected, and packetized into M packets.

CoopNet MDC System Architecture

Configuring MDC  GOF duration G = 1 second  M = 16 descriptions  Packet size P = 1250 bytes  T = 8 trees

Performance evaluation

Impact of Number of Distribution Trees

Effectiveness of MDC  Probability distribution of descriptions received vs. number of distribution trees  Root out-degree = 100  Maximum client out-degree = 4

Impact of Repair Time

Related work

Related Work  Application-level Multicast  Source Coding and Path Diversity

References  V. N. Padmanabhan, H. J. Wang, and P. A. Chou. Resilient Peer-to-Peer Streaming. Technical Report MSR-TR , Microsoft Research, Redmond, WA, March  V. N. Padmanabhan, H. J. Wang, P. A. Chou, and K. Sripanidkulchai. Distributing Streaming Media Content Using Cooperative Networking. In Proc. NOSSDAV, May 2002.

Q & A

Thank you!