Scheduling with Machine-Learning-Based Flow Detection for Packet-Switched Optical Data Center Networks

Wang, Lin; Wang, Xinbo; Tornatore, Massimo; Kim, Kwang Joon; Kim, Sun Me; Kim, Dae-Ub; Han, Kyeong‐Eun; Mukherjee, Biswanath

doi:10.1364/jocn.10.000365

Cited by 40 publications

(13 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a result, accuracy above 98% was obtained. Authors in [24] used ML classification techniques such as decision tree and naïve Bayes discretization to classify traffic flows into mouse flows (occur frequently but carry a small number of bytes) and elephant flows (occur occasionally but have a huge number of bytes). The paper presented classifiers performance in terms of accuracy and classification speed.…”

Section: Related Workmentioning

confidence: 99%

Machine Learning Classification and Regression Approaches for Optical Network Traffic Prediction

2021

View full text Add to dashboard Cite

Rapid growth of network traffic causes the need for the development of new network technologies. Artificial intelligence provides suitable tools to improve currently used network optimization methods. In this paper, we propose a procedure for network traffic prediction. Based on optical networks’ (and other network technologies) characteristics, we focus on the prediction of fixed bitrate levels called traffic levels. We develop and evaluate two approaches based on different supervised machine learning (ML) methods—classification and regression. We examine four different ML models with various selected features. The tested datasets are based on real traffic patterns provided by the Seattle Internet Exchange Point (SIX). Obtained results are analyzed using a new quality metric, which allows researchers to find the best forecasting algorithm in terms of network resources usage and operational costs. Our research shows that regression provides better results than classification in case of all analyzed datasets. Additionally, the final choice of the most appropriate ML algorithm and model should depend on the network operator expectations.

show abstract

Section: Related Workmentioning

confidence: 99%

Machine Learning Classification and Regression Approaches for Optical Network Traffic Prediction

2021

View full text Add to dashboard Cite

show abstract

“…The most commonly used architecture in current DCNs is the three-tier architecture, comprising a core layer, an aggregate layer, and an access layer, from top to bottom, as shown in Fig. 1 [3], [2], [4]. When DCNs process applications' requests, the requests first arrive at the core layer.…”

Section: Introductionmentioning

confidence: 99%

Performance Analysis of a Graph-Theoretic Load Balancing Method for Data Centers

AlShammari¹,

Mohammed²

2020

IJACSA

View full text Add to dashboard Cite

Modern data centers can process a massive amount of data in a short time with minimal errors. Data center networks (DCNs) use equal-cost, multi-path topologies to deliver split flows across alternative paths between the core layer and hosted servers, which could lead to significant overload if path scheduling is inefficient. Thus, distributing incoming requests among these paths is crucial for providing higher throughput and protection against link or switch failures. Several approaches have been proposed for path selection, mainly relying on round-robin and least-congested methods. In this paper, we propose a loadbalancing method based on betweenness centrality to improve the overall performance of a data center in terms of throughput, delay, and energy consumption. For evaluation, we compare our method with baseline methods of different DCN topologies: fattree, DCell, and BCube. On average, the evaluation results show that our method outperforms the others. It increases throughput by 202% and 33% while reducing delay by 20% and 22%, and energy consumption by 40% and 41% compared to the roundrobin and least-congested methods, respectively.

show abstract

“…It is widely used for wide‐scale operations, including the processing of large datasets. Also, DCNs require high‐performance modules to simultaneously handle a wide range of network services, entire communications, and data storage and transport 1‐4 . Therefore, using a software‐defined network (SDN) enhances DCN flexibility by decoupling the control core from forwarding functions.…”

Section: Introductionmentioning

confidence: 99%

BL‐Hybrid: A graph‐theoretic approach to improving software‐defined networking‐based data center network performance

AlShammari

Alenazi

2020

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Existing data centers can handle a massive amount of data transmitted in a short time with minimum errors. Data center networks (DCNs) use cost‐effective, multipath topologies to distribute flows through alternative paths between core layers and hosting servers. Software‐defined networking (SDN) separates the control plane from the data plane to ease network management. To perform active load balancing among different paths, we address the challenges of load balancing in multipath DCNs. However, most DCNs' load balancing systems rely on a static approach that distributes flows among different paths sequentially, regardless of the path status. Consequently, some paths become congested, degrading the overall performance of the DCN in terms of throughput and delay. Increasing the number of requests worsens the congestion problem, causing some connections to become overwhelmed, thereby decreasing DCN efficiency. In this paper, we propose and evaluate the performance of a novel dynamic load‐balancing system named BL‐Hybrid, which uses the graph‐theoretical centrality of betweenness as a base metric for data forwarding. We perform a comprehensive comparative analysis of our proposed method relative to the round‐robin and least‐congested algorithms in different traffic scenarios, using the most popular SDN‐based DCNs, namely, fat tree, BCube, and DCell. Our evaluation results show that BL‐Hybrid outperforms the round‐robin and least‐congested schemes. It maximizes the average throughput by 4% and 11% and the average data transferred by 7% and 13%, respectively. Also, it reduces the average round‐trip time latency by 14% and 41%, respectively.

show abstract

Scheduling with Machine-Learning-Based Flow Detection for Packet-Switched Optical Data Center Networks

Cited by 40 publications

References 13 publications

Machine Learning Classification and Regression Approaches for Optical Network Traffic Prediction

Machine Learning Classification and Regression Approaches for Optical Network Traffic Prediction

Performance Analysis of a Graph-Theoretic Load Balancing Method for Data Centers

BL‐Hybrid: A graph‐theoretic approach to improving software‐defined networking‐based data center network performance

Contact Info

Product

Resources

About