DCT${^2}$Gen: A Versatile TCP Traffic Generator for Data Centers

TL;DR

DCT2Gen is a configurable, realistic TCP traffic generator for data centers that models flow-level traffic and inter-rack differences, aiding research with more accurate simulations.

Contribution

It introduces the first publicly available, highly configurable traffic generator that models realistic data center traffic at the flow level, including intra- and inter-rack differences.

Findings

Accurately reproduces intra- and inter-rack traffic characteristics.

Uses six probability distributions for realistic traffic modeling.

Highly configurable for different data center sizes and types.

Abstract

Only little is publicly known about traffic in non-educational data centers. Recent studies made some knowledge available, which gives us the opportunity to create more realistic traffic models for data center research. We used this knowledge to create the first publicly available traffic generator that produces realistic traffic between hosts in data centers of arbitrary size. We characterize traffic by using six probability distribution functions and concentrate on the generation of traffic on flow-level. The distribution functions are described as step functions, which makes our generator highly configurable to generate traffic for different kinds of data centers. Moreover, in data centers, traffic between hosts in the same rack and hosts in different racks have different properties. We model this phenomenon, making our generated traffic very realistic. We carefully evaluated our approach and conclude that it reproduces these characteristics with high accuracy.