Methods for Predicting Behavior of Elephant Flows in Data Center Networks

AlAwadi, Aymen and Zaher, Maiass and Molnár, Sándor (2019) Methods for Predicting Behavior of Elephant Flows in Data Center Networks. INFOCOMMUNICATIONS JOURNAL, 11 (3). pp. 34-41. ISSN 2061-2079


Download (1MB) | Preview


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.

Item Type: Article
Subjects: H Social Sciences / társadalomtudományok > HD Industries. Land use. Labor / ipar, földhasználat, munkaügy > HD1 Industries / ipar > HD15 Risk management / kockázatkezelés
Q Science / természettudomány > QA Mathematics / matematika > QA75 Electronic computers. Computer science / számítástechnika, számítógéptudomány
Z Bibliography. Library Science. Information Resources / könyvtártudomány > Z665 Library Science. Information Science / könyvtártudomány, információtudomány
Depositing User: MTMT SWORD
Date Deposited: 06 Apr 2020 07:29
Last Modified: 06 Apr 2020 07:29

Actions (login required)

Edit Item Edit Item