1 / 15

GRAPES: a Generic Environment for P2P Streaming

GRAPES: a Generic Environment for P2P Streaming. Luca Abeni, Csaba Kiraly, Alessandro Russo, Marco Biazzini, Renato Lo Cigno. GRAPES. Generic Resource Aware P2P Environment for Streaming An open-source toolkit Simplifies implementation of P2P streaming solutions Does not impose

lilian
Download Presentation

GRAPES: a Generic Environment for P2P Streaming

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. GRAPES: a Generic Environmentfor P2P Streaming Luca Abeni, Csaba Kiraly, Alessandro Russo, Marco Biazzini, Renato Lo Cigno

  2. GRAPES • Generic Resource Aware P2P Environment for Streaming • An open-source toolkit • Simplifies implementation of P2P streaming solutions • Does not impose • Any approach • Any programming style

  3. Requirements • Portability • Written in C • No dependencies on external libraries, …

  4. Requirements • Neutrality • Should not impose any particular structure to the applications using it • No concurrency handling • No support for parallel activities • No code for receiving data from peers

  5. Requirements • Modularity • GRAPES functionalities are grouped into modules • Each module • Has its own API • Can be used independently from the others

  6. Communication • Applications based on GRAPES communicate through messages • Messages are sent sent by a network helper • Allows easy change of the protocol used for transmitting the messages

  7. Communication • GRAPES modules can • Directly send messages by invoking the network helper) • Simply construct them, leaving to the application the responsibility of sending them

  8. Overall Structure • GRAPES includes • A network helper • A Peer Set data type, to store information about peer’s neighbors • A Peer Sampling mechanism, providing each peer with continuously up- • to-date random samples of the entire population of peers; • A Chunk Trading module, allowing to send/receive pieces of a media • stream (called chunks); • A Chunk Buffer, used to store the received chunks so that they can be • forwarded to the other peers; • A Chunk ID Set data type, that can be used to send signalling informa- • tion about the received or needed chunks; • Scheduling functions

  9. Overall Structure • GRAPES includes • A network helper • A Peer Set data type, to store information about peer’s neighbors • A Peer Sampling mechanism, providing each peer with random samples of the entire population of peers • A Chunk Trading module, allowing to send/receive pieces of a media stream

  10. Overall Structure • A Chunk Buffer, used to store the received chunks so that they can be forwarded to the other peers • A Chunk ID Set data type, that can be used to send signaling information about the received or needed chunks; • Scheduling functions

  11. Peer Sampling • Most important functions include • Init() to initialize the peer sampling service • ParseData() to handle peer sampling messages coming from another peer is received) • AddNeighbour() to provide the ID of known peers (mainly used for bootstrapping) • GetNeighbourhood() to return a list of the known peers

  12. Signaling and chunk trading • GRAPES provides a chunk ID set • encodeChunkSignaling() allows the implementation of • buffermap messages • chunk offers (of one or more chunks) • chunk accepts (responses to an offer) • chunk requests (for one or more chunks) • chunk delivers (response to a request)

  13. The chunk buffer • Stores chunks received by an application • Most important functions are: • cb init() (to initialize a chunk buffer), • cb add chunk() (to insert a new chunk) • cb get chunks() (returns an ordered list of the chunks in the buffer)

  14. The chunk buffer • Different buffer management policies are possible: • Buffer discards chunks when a maximum size has been reached • Chunks are discarded when the difference between their playback time and the current time is too large • …

  15. Scheduling • Involves decisions such as • Selecting the chunk to be sent and the target peer in applications that periodically send chunks neighbours • Selecting a set of chunks to be requested from a neighbour, and that neighbour

More Related