Power efficient rate monotonic scheduling for multi-core systems

Min‐Allah, Nasro; Hussain, Hameed; Khan, Samee U.; Zomaya, Albert Y.

doi:10.1016/j.jpdc.2011.07.005

Cited by 32 publications

(55 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The communication amongst cores and main memory is achieved in two different ways: (i) through a single communication bus, which is also known as a shared memory model and (ii) through an interconnected network approach which is also known as a distributed memory model. The multi-core processor provides 4 times greater bandwidth, decreased energy consumption and running at lower frequency with the same voltage as a single core processor [56], [47]. Increasing the frequency of a single-core processor also increases the energy consumption due to the linear relationship of frequency to energy consumption [9].…”

Section: Multi-core Schedulingmentioning

confidence: 99%

“…To use multiple cores energy efficiently, the partitioning problem is an important concern. Multi-core hardware can increase the running application performance and minimize the energy consumption even more, if all the cores stay similarly active [47], [56]. If the cores are not equally loaded, it can waste CPU cycles and can increase message passing amongst different cores which reduces the running applications performance [50] and increases energy consumption.…”

Section: Multi-core Schedulingmentioning

confidence: 99%

“…In the last stage, designers needs to identify where tasks will execute. The load balancers [56] which are used to balance the workload amongst available servers in a compute cluster, to increase utilization level and minimize the chances of server overloading -such techniques are also used in multi-core processors. Apart from these, different techniques have been also implemented at processor level to save energy.…”

Section: Multi-core Schedulingmentioning

confidence: 99%

“…For example, processor speed controlling using DVFS (low utilization) [57] [58], a possibility of on/off switching (power cycling) [59] can save more energy but still considerable efforts are needed to find a conciliation between keeping a core idle, in on state or switch it off [60]. The deadline of a job must also be kept in mind, when scaling down the frequency or voltage of the core, as the job might miss the deadline due to increase in execution time while running on lower frequency [56]. Switching on/off a core also cost more energy (reconfiguration cost) and can degrade the system performance which increases energy consumption.…”

Section: Multi-core Schedulingmentioning

confidence: 99%

See 3 more Smart Citations

Energy efficient computing, clusters, grids and clouds: A taxonomy and survey

Zakarya

Gillam

2017

Sustainable Computing: Informatics and Systems

View full text Add to dashboard Cite

Section: Multi-core Schedulingmentioning

confidence: 99%

Section: Multi-core Schedulingmentioning

confidence: 99%

Section: Multi-core Schedulingmentioning

confidence: 99%

Section: Multi-core Schedulingmentioning

confidence: 99%

See 2 more Smart Citations

Energy efficient computing, clusters, grids and clouds: A taxonomy and survey

Zakarya

Gillam

2017

Sustainable Computing: Informatics and Systems

View full text Add to dashboard Cite

“…It is worth of mentioning, that the optimal utilization of the computational power is an general issue in various scientific and industrial applications, therefore advanced tools and method are developed and utilized [21][22][23].…”

Section: Measurement Setupmentioning

confidence: 99%

The implementation and the performance analysis of the multi-channel software-based lock-in amplifier for the stiffness mapping with atomic force microscope (AFM)

Sikora¹,

Bednarz²

2012

Bulletin of the Polish Academy of Sciences: Technical Sciences

View full text Add to dashboard Cite

Abstract. In this paper the implementation of the surface stiffness mapping method with the dynamic measurement mode of atomic force microscopy (AFM) is presented. As the measurement of the higher harmonics of the cantilever's torsional bending signal is performed, we are able to visualize non-homogeneities of the surface stiffness. In order to provide signal processing with the desired sensitivity and selectivity, the lock-in amplifier-based solution is necessary. Due to the presence of several useful frequencies in the signal, the utilization of several simultaneously processing channels is required. Therefore the eight-channel software-based device was implemented. As the developed solution must be synchronized with the AFM controller during the scanning procedure, the real-time processing regime of the software is essential. We present the results of mapping the surface stiffness and the performance tests results for different working conditions of the developed setup.

show abstract

Energy efficient genetic‐based schedulers in computational grids

Kołodziej

Khan

Wang

et al. 2012

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

In today's highly parametrized distributed computational environments, such as green grid clusters and clouds, the growing power and cooling rates are becoming the dominant part of the users' and system managers' budgets. Computational grids, owing to their sheer sizes, still require advanced methodologies and strategies for supporting the scheduling of the users' tasks and applications to the distributed resources. The efficient resource allocation becomes even more challenging when energy utilization, beyond the conventional scheduling criteria, such as Makespan, is treated as first-class additional scheduling objective. In this paper, we address the independent batch scheduling in computational grid as a bi-objective global minimization problem with Makespan and energy consumption as the main criteria. We apply the dynamic voltage and frequency scaling model for the management of the cumulative power energy utilized by the grid resources. We develop three genetic algorithms as energy-aware grid schedulers, which were empirically evaluated in three grid size scenarios in static and dynamic modes. The simulation results confirmed the effectiveness of the proposed genetic algorithm-based schedulers in the reduction of the energy consumed by the whole system and in dynamic load balancing of the resources in grid clusters, which is sufficient to maintain the desired quality level(s). systems must provide a wide range of services, high performance computing platforms, and process the large amounts of data. Moreover, a user in one locality might not be able to have control over other parts of the system. Therefore, to achieve full knowledge of the system and users profiles may be difficult, and the growing power and cooling rates are becoming the dominant part of the users' and system managers' budgets [3]. A severe increase in energy consumption in CGs may be the result of a disproportion of resource availability and resource provisioning in the light of computational demands [4]. All of the aforementioned issues will necessitate the development of intelligent resource management and scheduling techniques in grid computing. The modern grid schedulers should be able to optimize the conventional scheduling objectives simultaneously, such as Makespan, Flowtime, and Resource Utilization [5], and the new scheduling criteria, such as security requirements and energy consumed by all system's entities [6]. Energy-efficient scheduling in CGs becomes a complex endeavor owing to the multi-constraints and different optimization criteria and different priorities of the resource owners. Heuristic approaches seem to be the effective methodologies for designing energy-aware grid schedulers by trading off among various preferences and goals of the grid users, resource, service managers, and resource owners.In this paper, we address the problem of independent batch job scheduling in CGs, where tasks are processed in a batch mode and there are no dependencies among them. This scheduling scenario is very useful in illustrating many rea...

show abstract

Power efficient rate monotonic scheduling for multi-core systems

Cited by 32 publications

References 21 publications

Energy efficient computing, clusters, grids and clouds: A taxonomy and survey

Energy efficient computing, clusters, grids and clouds: A taxonomy and survey

The implementation and the performance analysis of the multi-channel software-based lock-in amplifier for the stiffness mapping with atomic force microscope (AFM)

Energy efficient genetic‐based schedulers in computational grids

Contact Info

Product

Resources

About