Efficient multicast data transfer with congestion control using dynamic source channels

TL;DR

This paper introduces a sequencer for efficient multicast data transfer that optimally maps application data to dynamic groups, improving transfer speed and fairness with minimal overhead.

Contribution

It presents a novel sequencer for dynamic multicast groups that enhances data transfer efficiency and fairness across multiple applications.

Findings

Sequencer achieves optimal behavior in data mapping.

File transfer efficiency is close to TCP with minimal overhead.

Download time remains nearly constant regardless of receiver number.

Abstract

The most efficient receiver-driven multicast congestion control protocols use dynamic channels. This means that each group has a cyclic rate variation with a continuously decreasing phase. Despite promising results in terms of fairness, using efficiently these dynamic groups could be a challenging task for application programmers. This paper presents a sequencer which maps out application data to dynamic groups in an optimal way. Multiple applications such as file transfer or video streaming, can use this sequencer, thanks to a simple API usable with any buffer containing the most important data first. To evaluate this solution, we designed a file transfer software using a FEC encoding. Results show the sequencer optimal behavior and the file transfer efficiency, as a single download generates only little more overhead than TCP . Moreover, download time is almost independent of the number of receivers, and is already faster than TCP with 2 competing downloads.