Methods for Predicting Behavior of Elephant Flows in Data Center Networks

TL;DR

This paper introduces a stochastic model to evaluate and compare the performance of flow scheduling algorithms in data center networks, focusing on their ability to handle elephant flows under various conditions.

Contribution

It presents a new stochastic performance evaluation model and provides a comprehensive statistical analysis of ECMP, Hedera, and DCTCP algorithms in data center environments.

Findings

Hedera remains risky for elephant flows due to throughput instability.

DCTCP struggles under high load scenarios.

The model aids in understanding flow behavior and algorithm effectiveness.

Abstract

Several Traffic Engineering (TE) techniques based on SDN (Software-defined networking) proposed to resolve flow competitions for network resources. However, there is no comprehensive study on the probability distribution of their throughput. Moreover, there is no study on predicting the future of elephant flows. To address these issues, we propose a new stochastic performance evaluation model to estimate the loss rate of two state-of-art flow scheduling algorithms including Equalcost multi-path routing (ECMP), Hedera besides a flow congestion control algorithm which is Data Center TCP (DCTCP). Although these algorithms have theoretical and practical benefits, their effectiveness has not been statistically investigated and analyzed in conserving the elephant flows. Therefore, we conducted extensive experiments on the fat-tree data center network to examine the efficiency of the algorithms under different network circumstances based on Monte Carlo risk analysis. The results show that Hedera is still risky to be used to handle the elephant flows due to its unstable throughput achieved under stochastic network congestion. On the other hand, DCTCP found suffering under high load scenarios. These outcomes might apply to all data center applications, in particular, the applications that demand high stability and productivity.