Cloud computing has aroused great interest from both academia and industry in the past decade. Due to Service Level Agreement(SLA), the cloud service providers have to guarantee the reliability of their systems. Howev...
详细信息
ISBN:
(纸本)9781509064151;9781509064144
Cloud computing has aroused great interest from both academia and industry in the past decade. Due to Service Level Agreement(SLA), the cloud service providers have to guarantee the reliability of their systems. However, unpredictable hardware or software failures are usually inevitable. For realtime tasks, system failure may cause them to miss their deadlines. Fault tolerance technique can efficiently avoid such ***-backup model is widely used to realize fault tolerance by duplicating a task into two copies – a primary copy and a backup copy. The redundancy introduced by backup copies incurs extra overhead for cloud systems. To address this issue, in this paper we propose a real-time fault-tolerant scheduling algorithm with rearrangement(RFTR) in cloud systems. Different from traditional scheduling algorithms, RFTR dynamically rearranges the execution orders of tasks. Simulation experiments are conducted on Cloud Sim platform. Compared with the existing scheduling algorithm FESTAL, RFTR shows excellent performance in both task schedulability and resource conservation.
Grid computing can integrate computational resources from different networks or regional areas into a high performance computational platform and be used to solve complex computing-intensive problems efficiently. Sche...
详细信息
ISBN:
(纸本)9780769533827
Grid computing can integrate computational resources from different networks or regional areas into a high performance computational platform and be used to solve complex computing-intensive problems efficiently. scheduling problem is an important issue in a grid computing environment, because of the heterogeneity of computing resources. This paper proposes an evolution-based dynamic scheduling algorithm (EDSA) for scheduling in grid computing environments. The proposed algorithm uses the genetic algorithm as search technique to find an efficient schedule in grid computing and adapts to variable numbers of computing nodes which has different computational capabilities. Furthermore, the hybrid crossover and incremental mutation operations within the algorithm can move the solution away from the local-optimal solution towards a near-optimal solution. And, a simulation with randomly generated task sets was performed to compare the performance with five other scheduling algorithms. The results show that the proposed EDSA outperformed all other schedulers across a range of scenarios.
This paper makes use of the feature of space channel of the massive MIMO which means that the channel of each user in the beam domain focus on part of the beam to propose a kind of multi-user scheduling algorithm base...
详细信息
ISBN:
(纸本)9789811065712;9789811065705
This paper makes use of the feature of space channel of the massive MIMO which means that the channel of each user in the beam domain focus on part of the beam to propose a kind of multi-user scheduling algorithm based on BDMA transmission. This algorithm takes the system and the rate as the principles to simplify the uplink and rate expression. The expression only needs to perform the calculation of the determinant according to the received related matrix of the base station. The dimension is only related to the quantity of the beam received by a single user. This paper obtains the instantaneous channel status information on the maximum beam of channel gain of each user through the estimation of the LS channel, also provides the transmission method selected by the beam, proposes to simplify the greedy algorithm for the user scheduling and at last perform the simulation directly at the multi user MIMO system user scheduling and characteristics of BRE which highlights the advantages of the algorithm.
In this paper, a hierarchical weighted round-robin scheduling (HWRR) algorithm is proposed in Load-Balanced two-stage switches. Following the change of traffic load, the proposed algorithm can adjust scheduling weight...
详细信息
ISBN:
(纸本)9781450364218
In this paper, a hierarchical weighted round-robin scheduling (HWRR) algorithm is proposed in Load-Balanced two-stage switches. Following the change of traffic load, the proposed algorithm can adjust scheduling weights and allocate bandwidth flexibility. Intensive dispatching will be performed when heavy load is detected. It can reduce the packet drop probability and meet the requirements of different priority traffic. The simulation results show that HWRR achieves stable performance on delay and cell drop probability, and performs well under several types of traffic.
Fault tolerance is an important aspect of real-time computer systems, since timing constraints must not be violated. For real-time single processor environment, Ghosh proposed two queue-based scheduling techniques. an...
详细信息
ISBN:
(纸本)0769509304
Fault tolerance is an important aspect of real-time computer systems, since timing constraints must not be violated. For real-time single processor environment, Ghosh proposed two queue-based scheduling techniques. an FSP (Feasible Shortest Path) algorithm and LTH(Linear lime Heuristics). Even though the FSP algorithm can produce optimal fault-tolerant schedules, it is not practical due to its time complexity The LTH algorithm is a greedy heuristics that closely approximates the optimal However since Ghosh's algorithm assumes that there is at most one fault within time interval df and does not consider interfault time it can deteriorate real-time scheduling performance due to unnecessary backup scheduling. In this paper;we have proposed an improved FSP algorithm on the more realistic assumption that there is no additional fault during minimum inter-fault time DeltaF after one fault occurs. The proposed algorithm can improve system performance by including more primary tasks in a fault-tolerant schedule and also reduce time complexity in generating backup schedules.
Long Term Evolution (LTE) is a main mobile communication technology in the future. For the reason that different kinds of multimedia applications have different quality of service (QoS) requirements, especially for Re...
详细信息
ISBN:
(纸本)9781467391948
Long Term Evolution (LTE) is a main mobile communication technology in the future. For the reason that different kinds of multimedia applications have different quality of service (QoS) requirements, especially for Real Time (RT) traffics, it is necessary to design an effective scheduling algorithm to satisfy the various kinds of QoS needs. In this paper we propose a new scheduling algorithm named DP-VT-MLWDF. The simulation results of the proposed scheduler indicate an evident improvement of the QoS performance parameters for RT traffics, e.g., packet loss rate, average throughput, fairness index.
Participants in the supply chain may have different information, leading to incomplete or inaccurate information when making decisions. To this end, a process and machine learning based collaborative scheduling algori...
详细信息
ISBN:
(纸本)9798400707032
Participants in the supply chain may have different information, leading to incomplete or inaccurate information when making decisions. To this end, a process and machine learning based collaborative scheduling algorithm for all materials is proposed. Design a health monitoring process for material supply chain based on R-tree dynamic indexing algorithm. Based on this, artificial neural networks in machine learning are applied to mine the data of the entire material supply chain. Through data mining, various data in the supply chain can be integrated and analyzed to improve information transparency and accuracy, and reduce information asymmetry. Adopting a dual layer scheduling model to achieve dual layer collaborative scheduling of materials. The experimental results show that the research method effectively improves the accuracy of data mining in the entire material supply chain, and the utilization rate of materials under this method is always higher than 95%.
The time-triggered principle establishes and maintains a system-wide synchronized time-base by the local clocks of the device. The proposed concept of Time-Triggered Ethernet (TTE) guarantees the determinacy of time-c...
详细信息
ISBN:
(纸本)9781509025237
The time-triggered principle establishes and maintains a system-wide synchronized time-base by the local clocks of the device. The proposed concept of Time-Triggered Ethernet (TTE) guarantees the determinacy of time-critical message for avionics and safety-critical systems. This paper presents an algorithm based on path-hop of tasks to generate a rational scheduling timetable to ensure is free of conflicts on the physical link. The scheduling timetable is periodic cycle by the Cluster Cycle as a unit. Cluster Cycle consists of several Integration Cycles (Min-cycle) with same time span. Time-Triggered (TT) flows are divided into several types by their path-hop parameter which can be obtained from path property. Different types of TT flows can be arranged into Cluster Cycle without conflict to achieve temporal determinacy. Meanwhile, the scheduling timetable unifies TT traffic and RC (Rate-Constrained) traffic defined in Avionics Full Duplex Switched Ethernet (AFDX) as well as BE (Best-Effort) traffic defined in ordinary Ethernet to improve bandwidth utilization. When all TT flows have been arranged into the scheduling timetable, the residual time resource can be used to transmit RC frames. Each vertex in the TTE network has its own scheduling timetable with different contents. Due to the clearly divided of timetable, it can support on-line modification for dynamical task insertion and withdrawal without congestion.
In order to improve the spectrum utilization while providing a fairness guarantee between users under the dynamic spectrum resources, the optimization problem for the scheduler of OFDM system with an unlimited queue c...
详细信息
ISBN:
(纸本)9781479960224
In order to improve the spectrum utilization while providing a fairness guarantee between users under the dynamic spectrum resources, the optimization problem for the scheduler of OFDM system with an unlimited queue capacity in cognitive radio networks (CRN) is first formulated. Then, under the constraints of the optimization problem, the Genetic algorithm for Cognitive Radio Scheduler (GACRS) is proposed based on the Simple Genetic algorithm (SGA) to solve the optimization problem. Simulation results show that, by adjusting the proportion factor in the fitness function, GACRS algorithm can guarantee the fairness among users while reaching the upper limit of the spectrum utilization.
The emergence of new data services in 3G networks is driving the support of advanced content-based billing mechanisms whose unique characteristics require that the OCS (Online Charging System) provide relative differe...
详细信息
ISBN:
(纸本)9781424423736
The emergence of new data services in 3G networks is driving the support of advanced content-based billing mechanisms whose unique characteristics require that the OCS (Online Charging System) provide relative differentiated services based on different contents. The OCS can be modeled by the PDD (Proportional Delay Differentiation) structure and it aims to maintain the prespecified mean system response-time delay ratios between different classes of requests. A novel request scheduling algorithm is proposed in this paper. The basic idea is that one of the requests from all classes of service queues is chosen with some priority probability which is based on the system queue-length, the request arrival rate, and the preassigned delay differentiation parameter. The experimental results show that this new algorithm significantly outperforms some other traditional PDD scheduling ones when request service delays increase and vary greatly from each other. Furthermore, it can efficiently meet 6 QoS requirements proposed in this paper.
暂无评论