A big video data transcoding service for social media over federated clouds
| Autor: | Maria Fazio, Alfonso Panarello, Antonio Puliafito, Antonio Celesti, Massimo Villari |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Service (systems architecture)
Multimedia Video transcoding Social media Cloud computing Federation Computer Networks and Communications Computer science 020207 software engineering Context (language use) Provisioning 02 engineering and technology Transcoding Video processing computer.software_genre Hardware and Architecture Scalability 0202 electrical engineering electronic engineering information engineering Media Technology Overhead (computing) Social media computer Software |
| Zdroj: | Multimedia Tools and Applications. 79:9037-9061 |
| ISSN: | 1573-7721 1380-7501 |
| DOI: | 10.1007/s11042-019-07786-9 |
| Popis: | Nowadays, the advent of social networks have revolutionised the traditional communication media. In recent years, the number of social media providers has rapidly grown. In this context, one of the major problems is the on-demand video streaming provisioning. In fact, more and more users require to post and access in real-time videos from anywhere in a short time. Therefore, a denial of service condition can cause for social media providers a loss of users and a consequent lose of money. Commonly, videos, before to be delivered, must be transcoded in order to fits both users’ hardware/software device and network capabilities, raising a big video data processing issue. In order to address such a concern, in this paper, we propose a Cloud federation system that enables social media providers to work together so as to take the advantages of a scalable video processing service. Experimental results demonstrate how the overhead due to setup and maintenance tasks of the federated environment is negligible compared to the benefits in terms of video transcoding performance. Moreover, we also demonstrate how Cloud federation can lighten and speed up the whole video processing service, by introducing an additional parallelization level. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink