Distributed Encoding Environment Based on Grids and IBP Infrastructure

HOLUB, Petr a Lukáš HEJTMÁNEK. Distributed Encoding Environment Based on Grids and IBP Infrastructure. In Selected Papers from TERENA Networking Conference 2004. TERENA, 2004, 10 s. ISBN 90-77559-04-3.

Základní údaje

• Originální název: Distributed Encoding Environment Based on Grids and IBP Infrastructure
• Autoři: HOLUB, Petr a Lukáš HEJTMÁNEK.
• Vydání: Selected Papers from TERENA Networking Conference 2004, 10 s. 2004.
• Nakladatel: TERENA

Další údaje

• Typ výsledku: Stať ve sborníku
• Utajení: není předmětem státního či obchodního tajemství
• WWW: Full paper. TNC2004 Proceedings.
• Organizační jednotka: Ústav výpočetní techniky
• ISBN: 90-77559-04-3
• Klíčová slova anglicky: Distributed Encoding Environment; multimedia processing; job scheduling; distributed storage
• Štítky: distributed encoding environment, distributed storage, job scheduling, multimedia processing

Anotace

This paper introduces an environment for distributed video transcoding based on Grid computing infrastructure and Internet Backplane Protocol storage infrastructure. A model for scheduling jobs with respect to location of data in distributed storage is introduced for optimal processing performance. We describe our abstraction library "libxio" providing a~uniform access interface to both local files and data stored on IBP. Modifications of video tools done based on this library in order to work with IBP are being presented as well as a~system that controls an overall distributed encoding process and helps scheduling system to schedule jobs in agreement with the distribution of data in IBP. Two pilot groups that use this system for preparing video content for streaming are mentioned.

Anotace anglicky

This paper introduces an environment for distributed video transcoding based on Grid computing infrastructure and Internet Backplane Protocol storage infrastructure. A model for scheduling jobs with respect to location of data in distributed storage is introduced for optimal processing performance. We describe our abstraction library "libxio" providing a~uniform access interface to both local files and data stored on IBP. Modifications of video tools done based on this library in order to work with IBP are being presented as well as a~system that controls an overall distributed encoding process and helps scheduling system to schedule jobs in agreement with the distribution of data in IBP. Two pilot groups that use this system for preparing video content for streaming are mentioned.