Background: Energy consumption is a major issue in Cloud Computing environments. Its efficient use can benefit in many ways such as cost saving, efficient utilization of resources and also saving the environment, as e...
详细信息
Background: Energy consumption is a major issue in Cloud Computing environments. Its efficient use can benefit in many ways such as cost saving, efficient utilization of resources and also saving the environment, as energy consumption, cost and time are important and decision making factors for both user as well as cloud service providers. QoS conscious scheduling of jobs along with energy awareness is very important, especially in cloud environment, where large datacenters are to be maintained and at the same time huge computations are involved. Optimal resource usage and price reduction are the direct and operational benefits for both users as well as the service providers. A substantial amount of energy is consumed by the underlying system resources. Hence, energy aware computations and scheduling is a big future concern that may heavily contribute to maintain the nature’s environmental systems, ecological balances and may avoid direct and indirect health hazards to all living beings. Omnidirectional benefits are the outcome of using energy aware scheduling techniques for Cloud environments and that too without compromising the Quality of Service. Methods: Software based scheduling and testing is done with DVFS (Dynamic Voltage and Frequency Scaling) based experiments for minimizing the processing cost, makespan time in Energy Aware environment so that in addition to energy saving, the Quality of Service is not compromised. Simulations are done using CloudSim with combinations of various Quality of Service (QoS) parameters along with the combinations of energy aware VM allocation policies. A comparison of these algorithms is shown with the normally used existing algorithms based on the Processing Cost, MakespanTime and Energy Utilization parameters. Results: A combination of Max-Min scheduling algorithm for cloudlet or task scheduling with Minimum Used Host scheduling algorithm for virtual machine allocation gives the most efficient environment in terms of Proces
In this paper, a buffer control strategy called limited greedy mode (LGM) is presented to handle fibre delay line (FDL) contentions which often occur in optical packet switch with shared FDL-buffer. In order to make f...
详细信息
In this paper, a buffer control strategy called limited greedy mode (LGM) is presented to handle fibre delay line (FDL) contentions which often occur in optical packet switch with shared FDL-buffer. In order to make full use of the FDL-buffer, three scheduling algorithms based on the LGM, i.e. Shortest Queue Based on Limited Greedy Mode (SQLGM), Increasing Queue Based on Limited Greedy Mode (IQLGM) and Shortest Queue Considering Output Void (SQCOV) are proposed. Performances of these three algorithms are evaluated by simulation experiments. Simulation results show that the SQCOV achieves the best performance. In addition, the buffering performance of shared FDL-buffer also depends on the greedy degree of LGM and the granularity of FDL. Copyright (C) 2009 John Wiley & Sons, Ltd.
Mobile Harbor is a movable floating structure with container loading/unloading equipment on board. Mobile Harbor is equivalent to a berth with a quay crane in a conventional port, except that it works with a container...
详细信息
In recent years, wireless networks have been widely adopted in industrial applications such as monitoring and control because of their flexibility and low cost. However, wireless networks use shared medium which requi...
详细信息
In recent years, wireless networks have been widely adopted in industrial applications such as monitoring and control because of their flexibility and low cost. However, wireless networks use shared medium which require appropriate Media Access Control (MAC) protocols to manage the shared medium. Time Division Multiple Access (TDMA)-based MAC protocols, which are collision free are predominantly used in industrial networks. To achieve the stringent requirements of reliability and timeliness in industrial applications, proper scheduling of the nodes accessing the shared channel is essential. A number of TDMA scheduling algorithms have been proposed in the literature but there is no study that have analyzed and compared the algorithms to help designers of new algorithms with basic information of the existing algorithms. The objective of this study is to provide basic findings that should be used by researchers looking for TDMA scheduling algorithms for industrial applications. To achieve the objectives of this article various categories of the TDMA scheduling algorithms that have been implemented in various industrial networks mostly Industrial Wireless Sensor Network (IWSN) which is widely deployed have been presented. Comparison of the various categories of the scheduling algorithms using reliability and delay as performance metrics which are crucial in industrial applications have been done. It was noted that hybrid TDMA scheduling algorithms have improved performance compared to distributed and centralized algorithms in random networks. Designing and testing of hybrid TDMA scheduling algorithms in the emerging industrial wireless networks such as LoRaWAN are recommended future work.
Wireless mesh network (WMN) is emerging as an important networking architecture for future wireless communications. The mesh mode supported in IEEE 802.16 protocol provides a TDMA solution for WMN, in which scheduling...
详细信息
MEMS-based storage is being developed as a new storage media that has several salient characteristics such as high-parallelism, high density, and low-power consumption. Because physical structures of MEMS-based storag...
详细信息
MEMS-based storage is being developed as a new storage media that has several salient characteristics such as high-parallelism, high density, and low-power consumption. Because physical structures of MEMS-based storage is different from those of hard disks, new software management techniques for MEMS-based storage are needed. Specifically, MEMS-based storage has thousands of parallel-activating heads, which requires parallelism-aware request scheduling algorithms to maximize the performance of the storage media. In this paper, we compare various versions of I/O scheduling algorithms that exploit high-parallelism of MEMS-based storage devices. Trace-driven simulations show that parallelism-aware algorithms can be effectively used for high capacity mass storage servers because they perform better than other algorithms in terms of the average response time when the workload intensity becomes heavy.
Task scheduling algorithms are key techniques in task management system of computing grid. Because of the uncertainty nature of a grid, traditional task scheduling algorithms do not work well in an open, heterogeneous...
详细信息
Providing quality of service (QoS) in wireless communication networks has become an important consideration for supporting variety of applications. IEEE 802.16 based WiMAX is the most promising technology for broadban...
详细信息
In this paper, two improved recursive maximum expansion scheduling algorithms for LTE uplink SCFDMA are proposed. The proposed Improved Recursive Maximum Expansion (IRME) scheme can achieve higher degree of freedom in...
详细信息
An optical overlaid star network is a class of agile all-photonic networks that consists of one or more core node(s) at the center of the star network and a number of edge nodes around the core node. In this architect...
详细信息
暂无评论