In this paper, we propose a new scheduling algorithm with economic theory, called Black Scholes Market (BSIII) algorithm for a class of Dynamic Jobs (DJ). BSM is based on the classic option pricing theory in investmen...
详细信息
ISBN:
(纸本)9780769526942
In this paper, we propose a new scheduling algorithm with economic theory, called Black Scholes Market (BSIII) algorithm for a class of Dynamic Jobs (DJ). BSM is based on the classic option pricing theory in investment- Black Scholes Pricing Model. The algorithm could meet the needs of dynamic flow jobs and select server to provide specific service through simulating an irrational market. Compared with Dynamic Weighted Round Robin (DWRR) and Dynamic Statistical Random (DSR) scheduling algorithms, BSM algorithm achieves a better performance in long time scheduling and the best average delay rate in different maximum job arrival rates. And from view of the stability, BSM is also much better than the other two algorithms.
In public Infrastructure-as-a-Service (IaaS), virtual machines, servers, storage, and network are provided by cloud service providers. As a cloud service provider, who is facing a task for time constraint, how to sche...
详细信息
ISBN:
(纸本)9781479951512
In public Infrastructure-as-a-Service (IaaS), virtual machines, servers, storage, and network are provided by cloud service providers. As a cloud service provider, who is facing a task for time constraint, how to schedule the service resources to achieve the lowest cost becomes more and more important. Recently, most of works about MapReduce task scheduling are focus on homogeneous MapReduce framework. In this paper, we present the ILP formulation for solving the MapReduce task scheduling for time constrains problem in heterogeneous environment. This method considers processing speed, energy cost and time constrains at the same time. By using the method, we can finish the task in time and achieving lowest energy cost. Then, we solve this problem efficiently by using genetic algorithm(GA). According to our experimental results, the ILP formulation we proposed can always achieve the best solution, it also reduced the energy consumption by 10.15% compared to genetic algorithm.
When a network scale for the traditional Ethernet is continuously expanded, an increase of the number of network hops will result in an augment of waiting delay uncertainty for a service transmission, which makes the ...
详细信息
ISBN:
(纸本)9781728121840
When a network scale for the traditional Ethernet is continuously expanded, an increase of the number of network hops will result in an augment of waiting delay uncertainty for a service transmission, which makes the network inefficiency and difficult to guarantee service quality. The time-triggered Ethernet adopts the global time synchronization and time triggering mechanism to ensure the service quality of high-priority services such as spaceflight tracking telemetry and control, network security, and vital signs monitoring, which possesses broad applications in aerospace, medical aid, and vehicle network. Currently, the time-triggered Ethernet triggers a time-triggered stream by employing an offline generated scheduling schedule, and is difficult to adapt to dynamic changes of a network, which leads to the network rigidity and inefficiency. Therefore, this paper proposes an algorithm for generating time-triggered schedules that can adjust time slots, flexibly. The time-triggered Ethernet prototype system is designed by utilizing the TrueTime, and the process of generating the scheduling table to trigger the time-triggered stream is simulated. Compared with the existing traditional algorithm, our algorithm can adjust the time slots according to the ratio of time-triggered streams to traditional traffic in the network, which can improve the time utilization rate by 30%.
Each Bluetooth device has different traffic loads and bandwidth requirements for various applications. In order to provide time-bounded and QoS-sensitive services with effective performance, this paper proposes a sche...
详细信息
ISBN:
(纸本)9781424410040
Each Bluetooth device has different traffic loads and bandwidth requirements for various applications. In order to provide time-bounded and QoS-sensitive services with effective performance, this paper proposes a scheduling algorithm adopting a resource reservation scheme that provides bandwidth guarantee to meet various end-to-end bandwidth requirements in scatternets. A coordination scheme is also proposed to efficiently manage guaranteed bandwidth services. In experiments, ns-2 simulator is used to demonstrate the QoS guarantee of the guaranteed bandwidth service.
In this paper, a novel concept is proposed to express the matching degree between available resources and Meta-tasks which uses weighted average Fuzzy Applicability. Furthermore, a new scheduling algorithm aimed at ti...
详细信息
ISBN:
(纸本)0769524869
In this paper, a novel concept is proposed to express the matching degree between available resources and Meta-tasks which uses weighted average Fuzzy Applicability. Furthermore, a new scheduling algorithm aimed at time and cost for Meta-tasks in Grid computing is designed on the basis of layered task scheduling model. The scheduling aim can be chosen according to user's requirement through setting different weights, for example, Time-Weight, Cost-Weight, etc. The comparison metric is the value of Fuzzy Applicability, and the optimum solution is the scheme which has the maximum. The experiment result shows that the algorithm can consider the factors of time and cost as a whole. It also ensures that, given the aim with different weights, the algorithm's performance is better than some existent heuristic algorithms.
ITU and other SDOs have launched oneM2M initiative recently and the standardization of M2M is now accelerating. The current access and core networks built for today's network services will be used as a common netw...
详细信息
ISBN:
(纸本)9789261140618;9781467346764
ITU and other SDOs have launched oneM2M initiative recently and the standardization of M2M is now accelerating. The current access and core networks built for today's network services will be used as a common network infrastructure for M2M network with some modifications. When the current access and core networks are used for both the current network services and M2M services, communications equipments at the network edge need to handle a large number of communication flows which are a mix of large volume data communication such as web access and M2M data communication at the same time. To satisfy the QoS requirements of many applications including M2M applications, communications equipments at the network edge will need to support both minimum guaranteed rate service and low delay forwarding service for small sized packets. In this paper, we propose a packet scheduling algorithm which can provide minimum guaranteed rate service and which can reduce the scheduling delay of small packets. It can be used in access network communications equipment such as edge router and OLT. We also evaluate the proposed algorithm by simulation.
Digital learning in Indonesian rural area faces some problems in delivering Video-on-Demand (VoD) as learning materials because of heterogeneous and limited network. Recently, data-driven overlay network (DONet) appro...
详细信息
ISBN:
(纸本)9781509016204
Digital learning in Indonesian rural area faces some problems in delivering Video-on-Demand (VoD) as learning materials because of heterogeneous and limited network. Recently, data-driven overlay network (DONet) approaches have attracted a lot of attention to solve VoD delivery problems in peer-to-peer based network. This paper presents an enhanced algorithm in overlay-network based digital learning built upon on previous research called CoolStreaming. The new algorithm takes into account bandwidth allocation of each node in scheduling content delivery process. The modified algorithm is then validated by comparing it with the previous algorithm. Validation of the proposed algorithm was based on simulation using OverSim and OMNET++ network simulator. We conducted three different scenarios to compare the proposed algorithm with the previous algorithm. The results show that the proposed algorithm has given better Quality of Service (QoS) performance than previous algorithm in all scenarios. The enhanced algorithm has indicated good quality in the investigated QoS parameters, namely delay and throughput.
As a new model of data storage and computing, cloud data center attracts much attention. Task scheduling is always a hot topic in the research of cloud data center. At the same time, As the cost of cloud data center i...
详细信息
ISBN:
(纸本)9781538662434
As a new model of data storage and computing, cloud data center attracts much attention. Task scheduling is always a hot topic in the research of cloud data center. At the same time, As the cost of cloud data center increases, the cost for users to use cloud data center becomes higher and higher. Less execution time means lower cost. Based on the above reasons, a short average execution time task scheduling algorithm is proposed in this paper, which can reduce the average execution time of the cloud data center according to the difference of the execution time of cloud tasks in the data center, and the effect of this method is more obvious with the increase of cluster size and the increase of the number of tasks. Finally, the algorithm is simulated in cloudsim to verify the effectiveness of the algorithm.
Long Term Evolution Advanced (LTE-A) is one of the fastest growing technologies used for transmitting data in cellular networks. It provides its subscribers with enhanced service capabilities and enhanced network perf...
详细信息
ISBN:
(纸本)9781538635810
Long Term Evolution Advanced (LTE-A) is one of the fastest growing technologies used for transmitting data in cellular networks. It provides its subscribers with enhanced service capabilities and enhanced network performance and this is carried through the smart deployment of new techniques and technologies. LTE-A is for improvement of the radio access part of cellular networks, for some time, it must co-exist with the 2G and 3G cellular networks, so interworking necessities, potential interference, resource management, etc. are an important issues. The Radio Resource Management (RRM) main role is to guarantee the efficient exploit of available radio resources using the available adaptation techniques, and to serve users based on their Quality of Service (QoS) parameters. In this paper, a novel dynamic neural Q-learning based scheduling algorithm is proposed for downlink transmission in LTE-A cellular network, aims to make a good trade-off between fairness and throughput. The proposed algorithm is based on the Q-learning technology and adoptable to variations in channel conditions. The key idea of our algorithm is how to intelligently choose the appropriate scheduling rule according to the predicted values of cell throughput and cell fairness index.
Forward link packet data service scheduling algorithm in mobile communication systems is researched in this paper. After analyzing a few kinds of familiar scheduling algorithms briefly, a novel scheduling algorithm is...
详细信息
ISBN:
(纸本)7563506861
Forward link packet data service scheduling algorithm in mobile communication systems is researched in this paper. After analyzing a few kinds of familiar scheduling algorithms briefly, a novel scheduling algorithm is proposed. Both the theoretical analysis and the simulation results show that this algorithm takes on several advantages. On the premise of satisfying the fairness criteria, it can increase the system throughput and decrease the average packet delay.
暂无评论