Scalable and Reliable Data Broadcast with Kascade

Abstract : Many large scale scientific computations or Big Data analysis require the distribution of large amounts of data to each machine involved. That distribution of data often has a key role in the overall performance of the operation. In this paper, we present Kascade, a solution for the broadcast of data to a large set of compute nodes. We evaluate Kascade using a set of large scale experiments in a variety of experimental settings, and show that Kascade: (1) achieves very high scalability by organizing nodes in a pipeline; (2) can almost saturate a 1~Gbit/s network, even at large scale; (3) handles failures of nodes during the transfer gracefully thanks to a fault-tolerant design.