scholarly journals Path based load balancing for data center networks using SDN

Author(s):  
V. Deeban Chakravarthy ◽  
B. Amutha

Due to the increase in the number of users on the internet and the number of applications that is available in the cloud makes Data Center Networking (DCN) has the backbone for computing. These data centre requires high operational cost and also experience the link failures and congestions often. Hence the solution is to use Software Defined Networking (SDN) based load balancer which improves the efficiency of the network by distributing the traffic across multiple paths to optimize the efficiency of the network. Traditional load balancers are very expensive and inflexible. These SDN load balancers do not require costly hardware and can be programmed, which it makes it easier to implement user-defined algorithms and load balancing strategies. In this paper, we have proposed an efficient load balancing technique by considering different parameters to maintain the load efficiently using Open FlowSwitches connected to ONOS controller.

2020 ◽  
Author(s):  
Hamid Nejadnik ◽  
Rasool Sadeghi ◽  
Sayed Mahdi Faghih Imani

Abstract Software Defined Networking (SDN) is a novel architecture that separates the data plane from the control plane using an external controller. Similar to traditional networks, load balancing has a great impact on the performance and availability of SDN. Therefore, the Controller Placement Problem (CPP) in SDN influences on the load balancing solutions. In this paper, various topologies of CPP including different load balancer controllers are simulated and evaluated in the SDN using the OFSwitch13 module of ns-3 network simulator. The results provide a solid comparison of the proposed topologies in different network situations.


Electronics ◽  
2021 ◽  
Vol 10 (22) ◽  
pp. 2820
Author(s):  
Syed Mohsan Raza ◽  
Shohreh Ahvar ◽  
Rashid Amin ◽  
Mudassar Hussain

Link failures frequently occur in communication networks, which negatively impacts network services delivery. Compared to traditional distributed networks, Software-Defined Networking (SDN) provides numerous benefits for link robustness to avoid services unavailability. To cope with link failures, the existing SDN approaches compute multiple paths and install corresponding flow rules at network switches without considering the reliability value of the primary computed path. This increases computation time, traffic overhead and end-to-end packets delay. This paper proposes a new approach called Reliability Aware Multiple Path Flow Rule (RAF) that calculates links reliability and installs minimum flow rules for multiple paths based on the reliability value of the primary path. RAF has been simulated, evaluated and compared with the existing approaches. The simulation results show that RAF performs better than the existing approaches in terms of computation overhead at the controller and reduces end-to-end packet delay and traffic overhead for flow rules installation.


2018 ◽  
Vol 10 (3) ◽  
pp. 157 ◽  
Author(s):  
Ramadhika Dewanto ◽  
Rendy Munadi ◽  
Ridha Muldina Negara

Equal Cost Multipath Routing (ECMP) is a routing application where all available paths between two nodes is utilized by statically mapping each path to possible traffics between source and destination hosts in a network. This configuration can lead to congestion if there are two or more traffics being transmitted into paths with overlapping links, despite the availability of less busy paths. Software Defined Networking (SDN) has the ability to increase the dynamicity of ECMP by allowing controller to monitor available bandwidths of all links in the network in real-time. The measured bandwidth is then implemented as the basis of the calculation to determine which path a traffic will take.  In this research, a SDN-based ECMP application that can prevent network congestion is made by measuring available bandwidth of each available paths beforehand, thus making different traffics transmitted on non-overlapped paths as much as possible. The proposed scheme increased the throughput by 14.21% and decreased the delay by 99% in comparison to standard ECMP when congestion occurs and has 75.2% lower load standard deviation in comparison to round robin load balancer.


2021 ◽  
Author(s):  
Ashvanth Kumar Selvakumaran

Software Defined Networking (SDN) technology has garnered much attention in the field of networking. Even though there have been several SDN based data centre (domain) implementations, there is a need to inter-connect multiple SDN domains. In this thesis we will focus on enabling inter-domain layer 2 switching. We propose an approach wherein a central controller is responsible for inter-domain switching while the domain controllers are responsible for intra-domain switching in their respective domains. To achieve this, the central controller initially communicates with the domain controllers to gather the overall topology of the network. From the overall topology, the central controller can derive the domain-level topology, compute the domain-level spanning tree and install the tree on the topology. In addition, the central controller also computes the inter-domain shortest path between any pair of domains. The shortest path information are then pushed to the domain controllers in order to setup the network-wide shortest path. We demonstrate the viability of the proposed approach by implementing it in OpenDayLight, a popular SDN platform. To further demonstrate the flexibility and openness of the approach, we have also successfully implemented a user case to achieve inter-domain load balancing.


2019 ◽  
Vol 28 (2) ◽  
pp. 298-339
Author(s):  
Dima Mansour ◽  
Haidar Osman ◽  
Christian Tschudin

AbstractLoad balancing is a mechanism to distribute client requests among several service instances. It enables resource utilization, lowers response time, and increases user satisfaction. In Named-Data Networking (NDN) and NDN-like architectures, load balancing becomes crucial when dynamic services are present, where relying solely on forwarding strategies can overload certain service instances while others are underutilized especially with the limited benefit of on-path caching when it comes to services. To understand the challenges and opportunities of load balancing in NDN, we analyze conventional load balancing in IP networks, and three closely related fields in NDN: congestion control, forwarding strategies, and data center management. We identify three possible scenarios for load balancing in NDN: facade load balancer, controller for Interest queues, and router-based load balancing. These different solutions use different metrics to identify the load on replicas, have different compliance levels with NDN, and place the load balancing functionality in different network components. From our findings, we propose and implement a new lightweight router-based load balancing approach called the communicating vessels and experimentally show how it reduces service response time and senses server capabilities without probing.


2018 ◽  
Vol 7 (2) ◽  
pp. 1-5
Author(s):  
Prabhjot Kaur ◽  
Jasmeen Kaur Chahal ◽  
Abhinav Bhandari

Software Defined Networking is an adaptable way of networking, which disconnects data forwarding plane and control-plane of system equipment’s and also solves issues in existing network infrastructure. More specifically, the control-plane of software defined network decides the advancing way of network flow with Centralized Control Manner (CCM). SDN (Software Defined Networking) is a strategy for making, planning and overseeing systems which intend to change this present unfortunate circumstance. It has been used in dissimilar areas, like a campus networks and data center systems. In this survey paper, we’ve reviewed the concept of (SDNs) Software Defined Networks, its architecture and applications. In the survey, it has been found that SDN load balancing has become more smart and efficient and reduces the statistic collection overhead and maintain better QoS (Quality of Service) data rates. In addition, we reviewed the direct routing based algorithms of Load Balancer and compare with Round Robin Strategy. Furthermore, we’ve reviewed and compared the existing work to get better idea about the concept of Load balancing.


The data center networks encompass various cloud services. Network congestion and network load imbalance may occur in data center networks due to elephant flows. In order to improve the throughput and overall utilization of the network, a dynamic load balancing mechanism has to be in place. Software Defined Networking (SDN) is used to perform the balancing of the network load. SDN can obtain the global view of the network and hence contain the status and topology of the entire data center network. The elephant flows can be split and send to multiple paths based on the current state of the network. The described idea is implemented in the OpenFlow environment and tested for improvement. The result shows the enhancement in throughput and network utilization.


2022 ◽  
Author(s):  
weimin gao ◽  
huang jiawei ◽  
Li zhaoyi ◽  
zou shaojun ◽  
wang jianxin

Abstract Modern data center topologies often take the form of a multi-rooted tree with rich parallel paths to provide high bandwidth. However, various path diversities caused by traffic dynamics, link failures and heterogeneous switching equipments widely exist in production data center network. Therefore, the multi-path load balancer in data center should be robust to these diversities. Although prior fine-grained schemes such as RPS and Presto make full use of available paths, they are prone to experi-ence packet reordering problem under asymmetric topology. The coarse-grained solutions such as ECMP and LetFlow effectively avoid packet reordering, but easily lead to under-utilization of multiple paths. To cope with these inefficiencies, we propose a load balancing mechanism called PDLB, which adaptively adjusts flowcell granularity according to path diversity. PDLB increases flowcell granularity to alleviate packet reordering under large degrees of topology asymmetry, while reducing flowcell granularity to obtain high link utilization under small degrees of topology asymmetry. PDLB is only deployed on the sender without any modification on switch. We evaluate PDLB through large-scale NS2 simulations. The experimental results show that PDLB reduces the average flow completion time by up to ∼11-53% over the state-of-the-art load balancing schemes.


Sign in / Sign up

Export Citation Format

Share Document