Survey on Traffic Management in Data Center Network: From Link Layer to Application Layer

Li, Weihe; Liu, Jingling; Wang, Shiqi; Zhang, Tao; Zou, Shaojun; Hu, Jinbin; Jiang, Wanchun; Huang, Jiawei

doi:10.1109/access.2021.3064008

Cited by 10 publications

(4 citation statements)

References 110 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The last decade has witnessed an ever-growing interest on optimizing Data Center Networks (DCN) given the critical services they run and the high CAPEX and OPEX these infrastructures entail. Among the large spectrum of traffic optimization mechanisms [24], Congestion Control (CC) has been especially explored in the literature [3,23,25,27,29,44,47]. Nowadays, most production DCNs run two main CC protocols: (𝑖) DCTCP [3], for datacenters based on the traditional TCP/IP stack, and (𝑖𝑖) DCQCN [47], for emerging RDMA-based DCNs.…”

Section: Introductionmentioning

confidence: 99%

MAGNNETO: A Graph Neural Network-Based Multi-Agent System for Traffic Engineering

Bernárdez¹,

Suárez-Varela²,

López³

et al. 2023

IEEE Trans. Cogn. Commun. Netw.

View full text Add to dashboard Cite

Current trends in networking propose the use of Machine Learning (ML) for a wide variety of network optimization tasks. As such, many efforts have been made to produce ML-based solutions for Traffic Engineering (TE), which is a fundamental problem in ISP networks. Nowadays, state-of-the-art TE optimizers rely on traditional optimization techniques, such as Local search, Constraint Programming, or Linear programming. In this paper, we present MAGNNETO, a distributed ML-based framework that leverages Multi-Agent Reinforcement Learning and Graph Neural Networks for distributed TE optimization. MAGNNETO deploys a set of agents across the network that learn and communicate in a distributed fashion via message exchanges between neighboring agents. Particularly, we apply this framework to optimize link weights in OSPF, with the goal of minimizing network congestion. In our evaluation, we compare MAGNNETO against several state-of-the-art TE optimizers in more than 75 topologies (up to 153 nodes and 354 links), including realistic traffic loads. Our experimental results show that, thanks to its distributed nature, MAGNNETO achieves comparable performance to state-of-the-art TE optimizers with significantly lower execution times. Moreover, our ML-based solution demonstrates a strong generalization capability to successfully operate in new networks unseen during training.

show abstract

Section: Introductionmentioning

confidence: 99%

MAGNNETO: A Graph Neural Network-Based Multi-Agent System for Traffic Engineering

Bernárdez¹,

Suárez-Varela²,

López³

et al. 2023

IEEE Trans. Cogn. Commun. Netw.

View full text Add to dashboard Cite

show abstract

“…Data centers have become pivotal for meeting enterprise computing demands, offering robust infrastructure for cloud computing services. They also facilitate intricate communications between a vast array of computing resources [1]. A recent industry report highlights that over the past five years, the U.S. data center market has seen an expenditure surge of $ 23 billion USD [2].…”

Section: Introductionmentioning

confidence: 99%

“…However, these models are generally inflexible. Specifically, they cannot adapt to accommodate an arbitrary number of nodes (or servers) 1 due to inherent structural limitations [12].…”

Section: Introductionmentioning

confidence: 99%

Advancing Data Center Networks: A Focus on Energy and Cost Efficiency

Chkirbene,

Hamila,

Al-Dweik

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Data centers serve as the backbone for cloud computing, enterprise services, and infrastructurebased offerings. One area of ongoing research in data center networking focuses on innovating new topologies for large-scale node connectivity. These topologies must incorporate fault-tolerant and efficient routing algorithms. Consequently, the data centers network topology must dynamically adapt to ever-changing application requirements. While traditional topology designs often emphasize scalability, they are typically limited by the necessity for dedicated switches to manage server connections. The development of software technologies that differentiate server and switch roles offers a unique opportunity to reconsider design priorities, paving the way for a more balanced assessment of scalability, energy efficiency, and infrastructure costs. Moreover, certain network topologies fail to be cost-effective due to their structural intricacies, often requiring far more node connections than those that are practically necessary. To address these challenges, we introduce VacoNet: a new flexible data center network topology that organizes nodes into structurally similar clusters, interconnected by a novel physical structure algorithm. Boasting high bisection bandwidth, VacoNet delivers robust network capacity, even when encountering bottlenecks. Furthermore, to connect a given set of nodes, VacoNet uses a minimal number of cables and switches, thereby drastically reducing both infrastructure costs and energy consumption. Simulation results show that VacoNet canreduce error rates by 20% and slashe infrastructure costs by 70% compared to existing solutions. Additionally, it performs tasks 30% faster, underscoring its superior performance.INDEX TERMS Data center, network topology, fexible connection, energy consumption, infrastructure cost.

show abstract

“…Other data center architectures such as B-cube and DCell also fulfill the high bandwidth use with suitable TCP flow control [5,6], and their packet controls on a flow are implemented into programmable network components such as OpenFlow switches and intelligent routers [7][8][9]. Despite adopting the excellent data center architecture, the developers of the data center network have suffered from a TCP incast problem, which leads the degradation of TCP throughput [10]. In detail, a network component has a queue for preserving TCP segment data (contained into a TCP flow), and shares it for multiple TCP flows.…”

Section: Introductionmentioning

confidence: 99%

Effective TCP Flow Management Based on Hierarchical Feedback Learning in Complex Data Center Network

Mizutani

2022

Sensors

View full text Add to dashboard Cite

Many studies focusing on improving Transmission Control Protocol (TCP) flow control realize a more effective use of bandwidth in data center networks. They are excellent ways to more effectively use the bandwidth between clients and back-end servers. However, these schemes cannot achieve the total optimization of bandwidth use for data center networks as they do not take into account the path design of TCP flows against a hierarchical complex structure of data center networks. To address this issue, this paper proposes a TCP flow management scheme specified a hierarchical complex data center network for effective bandwidth use. The proposed scheme dynamically controls the paths of TCP flows by reinforcement learning based on a hierarchical feedback model, which obtains an optimal TCP flow establishment policy even if both the network topology and link states are more complicated. In evaluation, the proposed scheme achieved more effective bandwidth use and reduced the probability of TCP incast up to 30% than the conventional TCP flow management schemes: Variant Load Balancing (VLB), Equal Cost Multi Path (ECMP), and Intelligent Forwarding Strategy Based on Reinforcement Learning (IFS-RL) in the complex data center network.

show abstract

Survey on Traffic Management in Data Center Network: From Link Layer to Application Layer

Cited by 10 publications

References 110 publications

MAGNNETO: A Graph Neural Network-Based Multi-Agent System for Traffic Engineering

MAGNNETO: A Graph Neural Network-Based Multi-Agent System for Traffic Engineering

Advancing Data Center Networks: A Focus on Energy and Cost Efficiency

Effective TCP Flow Management Based on Hierarchical Feedback Learning in Complex Data Center Network

Contact Info

Product

Resources

About