Worked on the issue of Energy Efficient solution and proposed 2 articles covering solution to NP-Hard Allocation problem by a novel machine learning-based metaheuristic algorithm which reduced runtime execution up to 88% and energy consumption up to 71%.
An artificial neural network-based approach for energy-efficient task scheduling in cloud data centers
Jan 9, 2020 – Sustainable Computing: Informatics and Systems (Elsevier)
Energy efficiency is considered a crucial objective in cloud data centers as it reduces cost and meets the standard set in green computing. Task scheduling an important problem becomes more complex and critical under energy efficiency consideration. Key issues in recent research on energy-efficient task scheduling are execution overhead and scalability. Machine learning has been widely employed for energy-efficient task scheduling problems but mostly used to predict resource consumption only instead of deciding the schedule itself. However, we used the neural network to decide which resource should be assigned to a given task independently. In this paper, we proposed an energy-efficient independent task scheduler using supervised neural networks with the aim to reduce makespan, energy consumption, execution overhead, and a number of active racks. The proposed artificial neural network-based scheduler takes an incoming task and current cloud environment state as input and predicts the best computing resource for the given task as output which compiles our aim. We used a genetic algorithm to generate a huge dataset (∼18 million training instances) and trained our neural network on this dataset using a backpropagation algorithm with 99.9% accuracy. We simulated experiments on heavily loaded and lightly loaded cloud environment and compared with well-known approaches: Genetic algorithm, MinMIN-MINMin heuristic, and Linear regression-based energy-efficient task schedulers. Results clearly indicate that the proposed work outperforms considered algorithms. In a heavily (lightly) loaded environment, it improves makespan by 59% (64%), energy consumption by 45% (71%), execution overhead by 88% (43%) respectively, and the number of active racks by 70%.
HIGA: Harmony-inspired genetic algorithm for rack-aware energy-efficient task scheduling in cloud data centers
Apr 19, 2019 – Engineering Science and Technology, an International Journal (Elsevier)
Reducing energy consumption in cloud data centers is one of the prime issues in the cloud community. It reduces energy-related costs and increases the lifespan of high-performance computing resources deployed in cloud data centers and it also helps in reducing carbon emissions. Along with energy efficiency, the problem of task scheduling is also one of the important problems considered in cloud data centers and it belongs to NP-class problems. With the energy consumption consideration, the problem of task scheduling becomes more complex to solve. In this work, we tackle the problem of energy-efficient task scheduling on modern cloud data center architecture and proposes a novel hybrid metaheuristic scheme harmony-inspired genetic algorithm (HIGA). It also addresses issues associated with metaheuristic algorithms. HIGA combines the exploration capability of genetic algorithm and exploitation capability of harmony search by which it intelligently senses local as well as a global optimal region without wasting time (iterations) in the local or global optimal region and provides quick convergence. Our primary objectives in this work are to reduce makespan and computing energy and secondary objectives are to reduce the energy consumed by the resources other than computing resources and reduce execution overhead associated with the scheduler. Collectively these objectives guide HIGA for better energy efficiency and performance while reducing the number of required resources (i.e. active racks). It indirectly also reduces cooling energy as we can switch off the rack components (air blower, cooling inlets) once racks become idle. Simulation analysis has been performed over independent task applications as well as real-world scientific applications like CyberShake, Epigenomics, and Montage. The result clearly manifests that the proposed HIGA provides up to 33% of energy savings and 47% of improvement in application performance (makespan) that too with 39% less execution overhead.
Link to article: https://doi.org/10.1016/j.jestch.2019.03.009
WGOA: Whale-Genetic Optimization Algorithm for Energy Efficient Task Scheduling in Cloud Data Center
Aug 5, 2018 – International Conference on Advanced Engineering Optimization Through Intelligent Techniques (Springer)
Energy-efficient task scheduling in heterogeneous cloud environments is now a day challenging problem and it belongs to NP-class problems. In this paper, we focused on minimizing energy consumption and improving performance (makespan) while reducing execution overhead for independent task scheduling. To achieve the same, we proposed a hybrid whale-genetic optimization algorithm to balance those two objectives; makespan and energy consumption. Our proposed algorithm also aims to reduce the execution time by reducing the number of iterations spend in the local optimal region. Extensive simulation has been carried out and compared over the original Whale Optimization Algorithm and well-known MinMin algorithm. Results clearly manifest that our proposed algorithm outperforms the original whale optimization algorithm as well as the MinMin heuristic algorithm and reduced execution overhead up to more than half of what required originally.
Link to article: https://www.springer.com/gp/book/9789811381959
Energy-aware Whale-optimized Task Scheduler in Cloud Computing
Dec 7, 2017 – 2017 International Conference on Intelligent Sustainable Systems (IEEE)
In cloud computing, one of the paramount problems is task scheduling and when we take the energy consumption of cloud data center into account it becomes more tedious and complex to solve. In this paper, we focused on minimizing the energy consumption along with makespan for independent task scheduling in cloud computing. For the same purpose, we used a whale optimization algorithm with a variant of the multi-objective model to take optimal scheduling decisions. Extensive simulation has been carried out and results clearly manifest that the purposed algorithm out-performs the well-known Min-Min heuristic for task scheduling in terms of energy consumption as well as makespan.
Link to article: https://ieeexplore.ieee.org/abstract/document/8389360