Dynamic energy-aware scheduling for parallel task-based application in cloud computing

Abstract

h i g h l i g h t s• Energy-aware run-time scheduler for task-based applications.• Model for estimating the application Energy consumption.• Methodology to automatically generate the required power consumption profile.• Multi-heuristic resource allocation algorithm to get solutions in polynomial time.• Energy saving/performance trade-off evaluation for different scenarios. a r t i c l e i n f o b s t r a c tGreen Computing is a recent trend in computer science, which tries to reduce the energy consumption and carbon footprint produced by computers on distributed platforms such as clusters, grids, and clouds. Traditional scheduling solutions attempt to minimize processing times without taking into account the energetic cost. One of the methods for reducing energy consumption is providing scheduling policies in order to allocate tasks on specific resources that impact over the processing times and energy consumption. In this paper, we propose a real-time dynamic scheduling system to execute efficiently taskbased applications on distributed computing platforms in order to minimize the energy consumption. Scheduling tasks on multiprocessors is a well known NP-hard problem and optimal solution of these problems is not feasible, we present a polynomial-time algorithm that combines a set of heuristic rules and a resource allocation technique in order to get good solutions on an affordable time scale. The proposed algorithm minimizes a multi-objective function which combines the energy-consumption and execution time according to the energy-performance importance factor provided by the resource provider or user, also taking into account sequence-dependent setup times between tasks, setup times and down times for virtual machines (VM) and energy profiles for different architectures. A prototype implementation of the scheduler has been tested with different kinds of DAG generated at random as well as on real task-based COMPSs applications. We have tested the system with different size instances and importance factors, and we have evaluated which combination provides a better solution and energy savings. Moreover, we have also evaluated the introduced overhead by measuring the time for getting the scheduling solutions for a different number of tasks, kinds of DAG, and resources, concluding that our method is suitable for run-time scheduling.

Task scheduling mechanisms in cloud computing: A systematic review

“…There are millions of users using the service, so the number of requests received to use the resources of the cloud is very large. 13,14 These requests have to be scheduled seeing the computation power, communication cost, task size, etc. Since we know that task scheduling is an NP-hard problem, we need certain algorithms for this allocation.…”

A prediction‐based and power‐aware virtual machine allocation algorithm in three‐tier cloud data centers

“…It also uses two VM selection policies for selecting VMs to migrate from an overloaded host. Juarez et al 13 proposed a real-time dynamic scheduling system to execute efficiently task-based applications on distributed computing platforms to minimize the energy consumption. The proposed algorithm minimizes a multiobjective function which combines the energy consumption and execution time according to the energy performance importance factor provided by the resource provider or user, also taking into account sequencedependent setup times between tasks, setup times and down times for virtual machines (VM), and energy profiles for different architectures.…”

A Bi-objective Scheduling Approach for Energy Optimisation of Executing and Transmitting HPC Applications in Decentralised Multi-cloud Systems

“…In this section, we limit our discussion to closely relevant work [16], [17], [18] that focuses on the energy consumption problem for scheduling HPC applications in cloud computing systems.…”

“…Their method clearly has associated overhead costs when the scheduling fails. On top of this, it has been demonstrated by Juarez et al [18] that creating or destroying VMs consumes nontrivial energy. Additionally, providing thresholds for the required frequency, in particular the maximum one, may limit the performance, which seems critical for deadlinebased applications.…”

“…A scheduling approach for optimising a bi-objective function of either energy consumption or makespan in heterogeneous cloud systems was proposed by Juarez et al [18]. They provide a combined cost function with a weighting factor α that indicates the user preference of either going for energy-efficiency or execution time.…”