ML-Based Online Traffic Classification for SDNs

Nsaif, Mohammed; Kovásznai, Gergely; Abboosh, Mohammed; Malik, Ali; Fréin, Ruairí de

doi:10.1109/citds54976.2022.9914138

Cited by 8 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As future work, it would be worth investigating how much ILP solvers scale for certain generalizations of the DCN model, such as using heterogeneous power consumption values for switches and links. In an ongoing work, we are upgrading the ILP model to save as much power as possible by adding more parameters, such as flow type [13]. Additionally, we are planning to experiment with pseudo-Boolean solvers as well.…”

Section: Discussionmentioning

confidence: 99%

Integer Programming Based Optimization of Power Consumption for Data Center Networks

Kovásznai,

Nsaif

2023

Acta Cybern

Self Cite

View full text Add to dashboard Cite

With the quickly developing data centers in smart cities, reducing energy consumption and improving network performance, as well as economic benefits, are essential research topics. In particular, Data Center Networks do not always run at full capacity, which leads to significant energy consumption. This paper experiments with a range of optimization tools to find the optimal solutions for the Integer Linear Programming (ILP) model of network power consumption. The study reports on experiments under three communication patterns (near, long, and random), measuring runtime and memory consumption in order to evaluate the performance of different ILP solvers.While the results show that, for near traffic pattern, most of the tools rapidly converge to the optimal solution, CP-SAT provides the most stable performance and outperforms the other solvers for the long traffic pattern. On the other hand, for random traffic pattern, Gurobi can be considered to be the best choice, since it is able to solve all the benchmark instances under the time limit and finds solutions faster by 1 or 2 orders of magnitude than the other solvers do.

show abstract

Section: Discussionmentioning

confidence: 99%

Integer Programming Based Optimization of Power Consumption for Data Center Networks

Kovásznai,

Nsaif

2023

Acta Cybern

Self Cite

View full text Add to dashboard Cite

show abstract

“…Finally, using ML algorithms to classify network traffic in real-time, SDN controllers can make more informed decisions about network traffic routing and resource allocation, leading to improved network performance and efficiency besides efficient power usage [34]. Overall, ML-based approaches have the potential to significantly reduce power consumption in DCNs, which is an important consideration for organizations looking to improve their energy efficiency and to reduce their carbon footprint.…”

Section: B Machine Learning-based Approachesmentioning

confidence: 99%

Survey of Routing Techniques-Based Optimization of Energy Consumption in SD-DCN

Nsaif,

Kovásznai,

Malik

et al. 2023

Infocommunications journal

View full text Add to dashboard Cite

The increasing power consumption of Data Center Networks (DCN) is becoming a major concern for network operators. The object of this paper is to provide a survey of state-of-the-art methods for reducing energy consumption via (1) enhanced scheduling and (2) enhanced aggregation of traffic flows using Software-Defined Networks (SDN), focusing on the advantages and disadvantages of these approaches. We tackle a gap in the literature for a review of SDN-based energy saving techniques and discuss the limitations of multi-controller solutions in terms of constraints on their performance. The main finding of this survey paper is that the two classes of SDNbased methods, scheduling and flow aggregation, significantly reduce energy consumption in DCNs. We also suggest that Machine Learning has the potential to further improve these classes of solutions and argue that hybrid ML-based solutions are the next frontier for the field. The perspective gained as a consequence of this analysis is that advanced ML-based solutions and multi-controller-based solutions may address the limitations of the state-of-the-art, and should be further explored for energy optimization in DCNs.

show abstract