Flow-Level Modeling of Parallel Download in Distributed Systems

Response time is the primary Quality of Service metric for parallel download systems, where pieces of large files can be simultaneously downloaded from several servers. Determining response times in such systems is still a difficult issue, because the way the network bandwidth is shared between flows is as yet not well understood. We address the issue by exploring the practical relevance of the hypothesis that flows share the network bandwidth according to the max- min fairness paradigm. We have implemented into a flow-level simulator a version of the algorithm, which calculates such a bandwidth allocation, which we have called the "progressive- filling flow-level algorithm" (PFFLA). We have programmed a similar model over NS2 and compared the empirical distri- butions resulting from both simulations. Our results indicate that flow-level predictions are very accurate in symmetric networks and good in asymmetric networks. Therefore, PFFLA would be extremely useful to build flow-level simulators and, possibly, to perform probabilistic QoS calculations in general P2P networks.