Scale and performance in a distributed file system
ACM Transactions on Computer Systems (TOCS)
Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment
Journal of the ACM (JACM)
Bistro: a framework for building scalable wide-area Upload applications
ACM SIGMETRICS Performance Evaluation Review
A taxonomy and survey of grid resource management systems for distributed computing
Software—Practice & Experience
Samsara: honor among thieves in peer-to-peer storage
SOSP '03 Proceedings of the nineteenth ACM symposium on Operating systems principles
Automatic measurement of quality metrics for colonoscopy videos
Proceedings of the 13th annual ACM international conference on Multimedia
A cooperative internet backup scheme
ATEC '03 Proceedings of the annual conference on USENIX Annual Technical Conference
Fine-grained device management in an interactive media server
IEEE Transactions on Multimedia
Modeling and verification of hybrid dynamic systems using multisingular hybrid Petri nets
Theoretical Computer Science
A resource scheduling approach for media uploading in video data center
PCM'12 Proceedings of the 13th Pacific-Rim conference on Advances in Multimedia Information Processing
| Hi-index | 0.00 |
This paper presents the design, implementation, and performance evaluation of a novel file uploading system. The system automatically uploads multimedia files to a centralized server given a client machine's hard deadline--the time when a client machine will exhaust its available storage space due to on-going recording of media files. If existing files have not been uploaded and removed from the client machine's hard disk by the deadlines, existing files may be overwritten or new files may not get created. Our uploading system was designed to provide a practical solution for emerging business needs. For instance, our system can be used in medical practice to gather videos generated from medical devices located in various procedure rooms for post-procedure analysis, and in law enforcement to collect video recordings from police cars during routine patrolling. Here we investigate two upload scheduling algorithms that determine which client machine should upload its file(s) first. We introduce two emergency control algorithms to handle situations when a client machine is about to exhaust its hard disk space. We evaluate the proposed algorithms via simulations and analysis. Our performance studies show that the upload scheduling algorithms and the emergency control algorithms have a significant impact on overall system performance.