Non-cooperative power and latency aware load balancing in distributed data centers

Tripathi, Rakesh; Vignesh, S.; Tamarapalli, Venkatesh; Chronopoulos, Anthony T.; Siar, Hajar

doi:10.1016/j.jpdc.2017.04.006

Cited by 32 publications

(26 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After experimentation, it is observed that it outperformed the other traditional algorithms. There are models available for load balancing between various data centers 18,19 . This is achieved by looking after the service type of the links.…”

Section: Literature Survey and Related Workmentioning

confidence: 99%

Intelligence‐enabled approach for load balancing in software‐defined data center networks

Fancy

Pushpalatha

2021

Int J Communication

View full text Add to dashboard Cite

Summary Software‐defined network provides eminent solutions for many complex network management functionalities in a data center network (DCN). One of the major tasks in any network is the load balancing in the available links. Due to dynamic data traffic nature in network, it is necessary to perform deep learning approaches for the long‐term and the short‐term data. This paper proposes a splitting policy‐based RL network (SPRLN) approach, a reinforcement learning‐based proactive load balancing algorithm that avoids the poll to the controller after the switch encounters an abnormality. The proposed method has been tested with simulations and found successful in improving the overall network performance by taking appropriate action for reward maximization. The testbed environment is treated as a Q‐learning algorithm; here, the optimality is defined as the path having the least score so that the overloaded path in that particular time can be avoided. An artificial neural network is needed because the data are uncertain all the time. Thus, the proposed SPRLN method yields 30% increased throughput and with 80% reduced data loss, when compared to existing approaches.

show abstract

Section: Literature Survey and Related Workmentioning

confidence: 99%

Intelligence‐enabled approach for load balancing in software‐defined data center networks

Fancy

Pushpalatha

2021

Int J Communication

View full text Add to dashboard Cite

show abstract

“…In recent times, the algorithmic game theory is a significant growth area in theoretical computer science. There are different game-theoretical approaches available like cooperative and non-cooperative games [21,22], symmetric and non-symmetric, simultaneous and sequential move games, constant sum, zerosum and non-zero-sum games. Stackelberg model is one of the strategic game models in which the leader firm makes its first move, and then the follower firm makes its next move sequentially [23].…”

Section: Related Workmentioning

confidence: 99%

Game theoretical approach for load balancing using SGMLB model in cloud environment

et al. 2020

View full text Add to dashboard Cite

On-demand cloud computing is one of the rapidly evolving technologies that is being widely used in the industries now. With the increase in IoT devices and real-time business analytics requirements, enterprises that ought to scale up and scale down their services have started coming towards on-demand cloud computing service providers. In a cloud data center, a high volume of continuous incoming task requests to physical hosts makes an imbalance in the cloud data center load. Most existing works balance the load by optimizing the algorithm in selecting the optimal host and achieves instantaneous load balancing but with execution inefficiency for tasks when carried out in the long run. Considering the long-term perspective of load balancing, the research paper proposes Stackelberg (leader-follower) game-theoretical model reinforced with the satisfaction factor for selecting the optimal physical host for deploying the tasks arriving at the data center in a balanced way. Stackelberg Game Theoretical Model for Load Balancing (SGMLB) algorithm deploys the tasks on the host in the data center by considering the utilization factor of every individual host, which helps in achieving high resource utilization on an average of 60%. Experimental results show that the Stackelberg equilibrium incorporated with a satisfaction index has been very useful in balancing the loading across the cluster by choosing the optimal hosts. The results show better execution efficiency in terms of the reduced number of task failures by 47%, decreased 'makespan' value by 17%, increased throughput by 6%, and a decreased frontend error rate as compared to the traditional random allocation algorithms and flow-shop scheduling algorithm.

show abstract

“…Subrata et al (2007) proposed a non-cooperative game-based framework to solve the allocation problem in computational grids. Tripathi et al (2017) considered the allocation problem in distributed data centers. Xiao et al (2017) considered interconnected systems and proposed a fairness aware resource allocation scheme using Q-learning.…”

Section: Related Workmentioning

confidence: 99%

An evolutionary approach for optimal multi-objective resource allocation in distributed computing systems

Kishor

Niyogi

2020

Concurrent Engineering

View full text Add to dashboard Cite

Resource allocation in a distributed computing system is the process of allocating the workload across multiple computing resources to optimize the required performance criteria. In this article, a resource allocation problem that arises in a distributed system consisting of multiple heterogeneous servers is addressed. The problem is modeled as a multi-objective problem with two conflicting objectives: (a) to minimize the users’ expected response time and (b) to reduce the utilization imbalance between servers. To satisfy these objectives simultaneously, first, both the objectives are considered in an integrated manner, and an optimization problem is formulated. Second, the optimization problem is cast into a game-theoretic setting and modeled as a non-cooperative game, called a non-cooperative resource allocation game. Finally, to solve the game, a differential evolution-based co-evolutionary framework (DECEF) is proposed. To evaluate the performance of DECEF, a rigorous simulation study is carried out. Furthermore, to assess the relative performance of DECEF, it is compared against two existing approaches, from various aspects, including system utilization, system heterogeneity, and system size. The experimental results show that DECEF provides better system-wide performance while optimizing both the objectives.

show abstract

Non-cooperative power and latency aware load balancing in distributed data centers

Cited by 32 publications

References 24 publications

Intelligence‐enabled approach for load balancing in software‐defined data center networks

Intelligence‐enabled approach for load balancing in software‐defined data center networks

Game theoretical approach for load balancing using SGMLB model in cloud environment

An evolutionary approach for optimal multi-objective resource allocation in distributed computing systems

Contact Info

Product

Resources

About