A speed-based adaptive dynamic parallel downloading technique
ACM SIGOPS Operating Systems Review
QoS Driven Parallelization of Resources to Reduce File Download Delay
IEEE Transactions on Parallel and Distributed Systems
Minimizing file download time in stochastic peer-to-peer networks
IEEE/ACM Transactions on Networking (TON)
Future Generation Computer Systems
A novel algorithm for fast retrival of information from a multiprocessor server
SEPADS'08 Proceedings of the 7th WSEAS International Conference on Software Engineering, Parallel and Distributed Systems
Accessing data from many servers simultaneously and adaptively in data grids
Future Generation Computer Systems
A heterogeneous peer-to-peer network testbed
ICUFN'09 Proceedings of the first international conference on Ubiquitous and future networks
Optimal File Splitting for Wireless Networks with Concurrent Access
NET-COOP '09 Proceedings of the 3rd Euro-NF Conference on Network Control and Optimization
Parallel state transfer in object replication systems
DAIS'07 Proceedings of the 7th IFIP WG 6.1 international conference on Distributed applications and interoperable systems
Characterizing the file hosting ecosystem: A view from the edge
Performance Evaluation
Parallel and multi-wavelength downloading in optical grid networks
Photonic Network Communications
Optimizing end-to-end throughput for data transfers on an Overlay-TCP path
NETWORKING'05 Proceedings of the 4th IFIP-TC6 international conference on Networking Technologies, Services, and Protocols; Performance of Computer and Communication Networks; Mobile and Wireless Communication Systems
| Hi-index | 0.01 |
In the presence of multiple mirror servers that can servethe same information, clients can improve their performancewhen downloading large files by conncurrently retrievingdifferent parts of the file from different servers. Inprevious work, experiments with a single client have shownthe potential for parallel downloading schemes to improvethe performance perceived by this single client. In this paperwe consider the question of what happens when paralleldownloading is widely adopted within the Internet. Tothat end we start with an experimental evaluation of variousparallel downloading schemes and their performance fromthe perspective of a single client. We use these results as abenchmark for a set of simulations in which we investigatethe performance of parallel downloading when performedby multiple clients. We find that, because parallel downloadingincurs additional overhead as a result of its use ofmore network connections, large-scale use of this techniquecan actually lead to overall degradation of the performanceexperienced by clients. Further, we uncover a "fairness"tradeoff in that any performance improvement experiencedby clients performing parallel downloading comes at the expenseof clients who simply go to a single server to retrievefiles.