A Speed-Based Adaptive Dynamic Parallel Downloading Technique
adPD：一种速度自适应的动态并行下载技术

In this paper, we describe adPD, an improved parallel downloading approach to retrieving files by estab- lishing parallel connections with multiple copies. adPD assigns each server equal portion of file as large as possible, which ensures client downloading from each server without interrupt. And after faster server finish its own work, ad- PD will reallocate part of unfinished work of some slower server to the faster server. By this means, adPD can dynam- ically adjust the proportion of a file retrieved from each server during the downloading process, and can also reduce the number of block requests. Experiment results show that adPD is an effective parallel download scheme especially in Peer-to-Peer environments. Our experiment results show that adPD is an effective parallel download scheme.