Energy-aware stochastic scheduling model with precedence constraints on DVFS-enabled processors

Sajid, Mohammad; Raza, Zahid

doi:10.3906/elk-1505-112

Cited by 14 publications

(9 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Being a bi‐objective problem, energy‐aware scheduling requires determining the solutions with maximum possible optimized makespan and energy consumption . Due to NP‐complete nature, a large number of energy‐aware heuristics and meta‐heuristics have been reported in the literature for different types of applications as well as execution environment …”

Section: Related Researchmentioning

confidence: 99%

See 1 more Smart Citation

Energy‐efficient quantum‐inspired stochastic Q‐HypE algorithm for batch‐of‐stochastic‐tasks on heterogeneous DVFS‐enabled processors

Sajid

Raza

2019

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

Scheduling on dynamic voltage and frequency scaling enabled processors to determine the Pareto-optimal solutions with optimized makespan and energy consumption demands faster multi-objective scheduling algorithms. In general, the problem of multi-objective optimization, ie, finding the Pareto-optimal solutions to optimize two or more QoS parameters, has been proven to be an NP-complete problem. In this work, we propose a novel energy-efficient quantum-inspired stochastic Q-HypE algorithm to schedule the batch-of-stochastic-tasks (BoT) on DVFS-enabled processors with the aim to optimize the makespan of BoT as well as the energy consumption of processors. The stochastic processing times of tasks are drawn from independent probability distributions. The proposed Q-HypE algorithm evolves from combined characteristics of quantum computing and a hypervolume based multi-objective optimization HypE algorithm. The proposed Q-HypE algorithm simultaneously minimizes the makespan and energy consumption of the Pareto-optimal solutions whereas the dynamics of quantum computing accelerate the process of HypE to further minimize the overheads of hypervolume estimation. Experimental results reveal the effectiveness of the proposed Q-HypE algorithm both in terms of the number and quality of solutions offered. KEYWORDS batch-of-stochastic-tasks, dynamic voltage and frequency scaling, HypE, Pareto optimality, quantum computing, stochastic scheduling INTRODUCTIONOn one hand, data centres are the backbone of the world economic system due to millions of critical financial/federal electronic transactions per second that leads to huge financial loss with 1-second downtime. On the other hand, data centres are the fastest-growing consumers of electrical energy. It has been reported in 2013 that 3 million data centres (12 million servers) in the US consumed an estimated 91 billion kilowatt-hours of electrical energy. Furthermore, it is expected by 2020 that this energy consumption will consume 140 billion kilowatt-hours of electrical energy, costing $13 billion monetarily and adding 150 million metric tons (MMT) carbon pollution annually. 1,2 Therefore, higher financial costs, high energy consumption and negative environmental impacts foster energy-aware data centres consisting of energy-efficient hardware as well as software resources. 3For energy-efficient management of resources, dynamic voltage and frequency scaling (DVFS)-enabled processors are the potential candidates that allow the supply voltage to change at run time. An energy-aware algorithm can adjust supply voltage and frequency at suitable times in order to offer an optimized consideration of performance and energy consumption. 4,5 In general, scheduling is concerned with the allocation of jobs to processors/cores with the objective to optimize one or more performance measures such as makespan, energy consumption, reliability, load balancing, and so on. 6 On DVFS-enabled processors, the scheduling algorithm encounters the challenge of selecting the best processor and then selec...

show abstract

Section: Related Researchmentioning

confidence: 99%

“…The performance of ESHEFT algorithm has been analyzed with SHEFT and ECS algorithms for randomly created jobs. ESHEFT algorithm performs better than peers …”

Section: Related Researchmentioning

confidence: 99%

Energy‐efficient quantum‐inspired stochastic Q‐HypE algorithm for batch‐of‐stochastic‐tasks on heterogeneous DVFS‐enabled processors

Sajid

Raza

2019

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…In order to reduce the impact of resource performance fluctuations on the deadline, the authors proposed to use the replication technique. In [17] the processing times of tasks were assumed to follow independent random variables. The authors proposed an energy-aware stochastic algorithm with the objective of optimizing execution time and energy consumption.…”

Section: Related Workmentioning

confidence: 99%

Task graph scheduling in the presence of performance fluctuations ofcomputational resources

Malakoutifar¹,

Motallebi²

2019

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Most of the existing work in the area of task graph scheduling considers resources with fixed processing capacity. The algorithms in these works rely on an estimation of the execution times of tasks on different resources. However, in practice, due to fluctuations in performance of cloud resources, these algorithms have challenges in these environments. In this paper, we focus on the problem of fault-tolerant scheduling of task graphs in the presence of performance fluctuations of computational resources. With the aim of reducing the adverse impacts of both soft errors and resource performance degradations, we propose an opportunistic task replication scheme that uses idle durations of resources for replicating tasks. Unlike the previous works, the proposed algorithm does not rely on estimation of task execution times for finding idle resources. We introduce the notion of concurrency graphs and propose a graph theory-based algorithm for finding the number of idle resources during the execution of a set of tasks. The appropriate redundancy for each task is chosen with respect to the number of idle resources and the characteristics of the set of tasks that are being processed concurrently. Simulation experiments show that, in most situations, the proposed algorithm outperforms the previous algorithms in terms of average execution time and cost.

show abstract

“…Sajid and Raza [14] proposed the ESHEFT algorithm to schedule the tasks in DVFS enabled heterogeneous multi-core systems for optimizing turnaround time and energy consumption. Kumar and Raghunathan [15] proposed a hybrid approach for better energy saving using dynamic power management (DPM) with sleep state and DVFS for virtualized server clusters.…”

Section: Non-preemptive Tasksmentioning

confidence: 99%

“…The hyper-planes exact test (HET) [19] provides reduced scheduling points based on Eq. (14). The USFI algorithm is applied to HET with reduced scheduling points.…”

Section: Reduced Scheduling Points Set For the Rate Monotonic Algorithmmentioning

confidence: 99%

Energy optimized scheduling for non-preemptive real-time systems

Kannaian¹,

Palanisamy²

2017

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Abstract:Embedded systems have geared up their applications in the field of smart phones, mobile devices, automobiles, and medical instrumentation. Along with the growth of technology, the functional requirements of the systems have also grown multifold. To accommodate these ever increasing functional requirements, the hardware should be designed to include the maximum capability, which in turn increases power consumption. Thus this has to be dynamically managed using software techniques. The dynamic voltage and frequency scaling (DVFS) feature provided in most of the recent processors has enabled power optimization through software. With this feature the processor operating frequency can be reduced on the fly through software. This paper proposes two approaches. In the first one, the hyper period-based method (HPBM), the slow down factor is calculated for the hyper period and the second method involves applying uniform slow down with frequency inheritance (USFI) with a hyper plane exact test (HET). Both approaches aim to calculate slow down factors for the tasks to be executed in the processors with the DVFS feature. The main objective of this work is to provide energy optimized scheduling as well as to minimize the overheads in slow down factor calculations.

show abstract

Energy-aware stochastic scheduling model with precedence constraints on DVFS-enabled processors

Cited by 14 publications

References 14 publications

Energy‐efficient quantum‐inspired stochastic Q‐HypE algorithm for batch‐of‐stochastic‐tasks on heterogeneous DVFS‐enabled processors

Energy‐efficient quantum‐inspired stochastic Q‐HypE algorithm for batch‐of‐stochastic‐tasks on heterogeneous DVFS‐enabled processors

Task graph scheduling in the presence of performance fluctuations ofcomputational resources

Energy optimized scheduling for non-preemptive real-time systems

Contact Info

Product

Resources

About