Load-adaptive inter-piconet scheduling in small-scale Bluetooth scatternets

Gang scheduling in multi-core clusters implementing migrations

The LTE specification provides Quality of Service (QoS) of multimedia services with fast connectivity, high mobility, and security. However, 3GPP specifications have not defined scheduling algorithms to support real-time and non-realtime application services, this has led to the development of a variety of scheduling algorithms in the recent years. Performance is a key factor for assessing the scheduling algorithms. However, the performance analysis and evaluation for existing scheduling algorithms are limited to QoS parameters (e.g. in terms of throughput, packet loss and delay) and/or fairness metrics. It is not clear how these scheduling algorithms perform in terms of Quality of Experience (QoE) metrics (e.g. MOS score) which are more closely linked with end user experience or user perceived quality. In this paper, we investigate and evaluate the performance of three popular LTE downlink schedulers (proportional fairness (PF), exponential proportional fairness (EXP-PF), and modified largest weight delay first (M-LWDF)) in terms of QoE metric (i.e. MOS score) in addition to normal QoS metrics (e.g. packet loss and delay) for VoIP applications. Simulation results (based on LTE-Sim) show that the best suitable downlink scheduling algorithm is MLWDF, which has short end-to end delay (less than 50ms) and allowing maximum number of user access (more than 50 users) at acceptable MOS score (MOS over 3.5). The widely used scheduling algorithm (PF) will increase its end-to-end delay over 200ms, when number of user access is over 30. This will limit its application for VoIP when user access number increases. This work would help to assist the design and development of QoE-driven scheduling algorithms for mobile multimedia applications in the future.

As the penetration level of plug-in hybrid electric vehicles (PHEVs) increases, the stress on the distribution system will become more pronounced. In this paper, we concentrate on PHEV charging in public infrastructures, such as shared parking lots in commercial office campuses and shopping malls. In such clustered scenarios, PHEV charging demand can vary significantly and stochastically with time, and the overall system performance is a result of a trade-off between the peak load on the distribution system and the consumer specified charging deadlines that are met successfully. However, unlike distributed residential charging, those scenarios can benefit from centralized algorithms to schedule the charging of vehicles. This paper considers the design of such scheduling algorithms. We show that through proper design of the scheduling algorithms, the parking lot owner can balance distribution system load with quality of charging service.

This paper proposes a novel class of wavelength scheduling algorithms in optical burst switching networks. The proposed wavelength scheduling algorithms process a batch of data bursts together instead of processing them one by one. When a control burst with a reservation request arrives to a batch scheduler, the scheduler waits for a small amount of time, called the acceptance delay, before deciding to accept or reject the reservation request. After the acceptance delay has passed, the scheduler processes all the reservation requests that have arrived during the acceptance delay, then it accepts the requests that will maximize the utilization of the wavelength channels. We describe an optimal batch scheduler that serves as an upper bound on the performance of batch scheduling algorithms. Furthermore, we introduce four novel heuristic batch scheduling algorithms. The bursts of the proposed algorithms is evaluated using a discrete-event simulation model. Simulation results suggest that batch schedulers could decrease the blocking probability by 25% compared the best previously known wavelength scheduling algorithm.

Previously proposed wavelength scheduling algorithms in optical burst switching networks process each reservation request individually and in a greedy manner. In this paper we propose a new family of wavelength scheduling algorithms that process a batch of reservation requests together instead of processing them one by one. When a control burst with a reservation request arrives to a free batch scheduler, the scheduler waits for a small amount of time, called the acceptance delay, before deciding to accept or reject the reservation request. After the acceptance delay has passed, the scheduler processes all the reservation requests that have arrived during the acceptance delay, then it accepts the requests that will maximize the utilization of the wavelength channels. We describe an optimal batch scheduler that serves as an upper bound on the performance of batch scheduling algorithms. Furthermore, we introduce two heuristic batch scheduling algorithms. The performance of the proposed algorithms is evaluated using a discrete-event simulation model. Simulation results suggest that batch schedulers could decrease the blocking probability by 25% compared to the best previously known wavelength scheduling algorithm.

With the advent of increasingly higher numbers of processors on-chip, task scheduling has become an important concern in system design, and research in this area has produced substantial and diversified knowledge. As a result, the efficient management and taping of this knowledge has become a concern in itself. This paper addresses the issue of how to effectively extract performance information about a scheduling algorithm in the context of a set of applications, by learning the association rules between the applications' attributes and the algorithms' performance metrics. The new methodology that is presented serves to both increase the designer's knowledge about a particular scheduling algorithm and compare algorithms.

Scheduling algorithms are fundamental components in the process of resource management in mobile communication networks with heterogeneous QoS requirements such as delay, delay jitter, packet loss rate or throughput. The random characteristics of the propagation environment and the use of complex physical layers in order to combat this random behavior further complicates the design of simple, efficient, scalable and fair scheduling algorithms. This paper presents the main criteria used in the design of scheduling algorithms for 3G/4G mobile communications networks and provides a survey of scheduling mechanisms proposed for use in TDMA and CDMA based systems.Keywordsscheduling4GGPSutility functions

Cloud Computing provides a Computing environment where different resources, infrastructures, development platforms and software are delivered as a service to customers virtually on pay per time basis. Low cost, scalability, reliability, utility-based computing are important aspects of cloud computing. Job scheduling is an essential and most important part in any cloud environment. With increasing number of users, Job scheduling becomes a strenuous task. Ordering the jobs by scheduler while maintaining the balance between quality of services (QoS), efficiency and fairness of jobs is quite challenging. Scheduling algorithms are implemented considering parameters such as throughput, resource utilization, latency, cost, priority, computational time, physical distances, performance, bandwidth, resource availability. Though there are different scheduling algorithms available in cloud computing, a very less comparative study has been done on performance of various scheduling algorithms with respect to above mentioned parameters. This paper aims at a comparative study of various types of job scheduling algorithms that provide efficient cloud services.

Scheduling algorithm design is a critical and challenging issue in multiuser wireless networks with dynamic flows. The well-known Queue-length-based MaxWeight (QMW) scheduling algorithm can achieve throughput-optimality if there only exist persistent flows that are long-lived with infinite traffic arrival. In this paper, we propose a head-of-line access delay (HAD)-based scheduling algorithm and show that it is throughput-optimal when the flows are dynamic, i.e., they are short-lived with finite data to transmit. HAD is an online algorithm and does not require prior knowledge of the statistics of the arrival traffic and channel information. We also develop the Markov analytic model to study system performance and reveal important properties of the proposed HAD scheduling algorithm. To reduce the complexity of the analysis, we further study two approximation methods corresponding to different arrival traffic intensity. Performance evaluation shows that the HAD scheduling algorithm can outperform the classic QMW and stabilize the system at the presence of flow-level dynamics. Compared to the other existing algorithms, HAD is practical to implement with a better delay performance.

The demand for reliable and energy-efficient low-power wireless network protocols is growing up steadily with the flourish of the emerging Internet of Things (IoT) applications. The Time Slotted Channel Hopping (TSCH), part of the IEEE 802.15.4 standard, represents the latest generation of low-power and highly reliable MAC protocols. It orchestrates the medium access according to a time-frequency communication schedule. Several scheduling algorithms have been introduced for TSCH. In contrast to centralized and distributed scheduling approaches, autonomous scheduling does not occur any signaling overhead and appears as a promising approach. In this study, we review autonomous TSCH scheduling algorithms, including static and recently proposed adaptive algorithms. Then, we investigate, on a real-world testbed, the performance of representative static autonomous scheduling schemes for TSCH protocol. Experimental results show that they suffer from high number of queued packets and buffers overflow, which increase the network delay and decreases the delivery ratio. Finally, we qualitatively compare and discuss the performance of recent adaptive autonomous TSCH scheduling algorithms, showing that research on adaptive autonomous TSCH scheduling still has open issues.

A simulation-based scheduling system is discussed which was developed for a semiconductor Backend facility. Apart from usual dispatching rules it uses heuristic search strategies for the optimisation of the operating sequences. In practice hereby multiple objectives have to be considered, e.g. concurrent minimisation of mean cycle time, maximisation of throughput and due date compliance. Because the simulation model is very complex and simulation time itself is not negligible, we emphasise to increase the convergence of heuristic optimisation methods, consequentially reducing the number of necessary iterations. Several realised strategies are presented.

To provide uninterrupted service to all subscribers in a wireless network, we need to incorporate a low cost, flexible Heterogeneous network which will be able to link with any kind of network for efficient spectrum utilization, hence improved system capacity.In this connection, Wi-Fi/ Wi MAX integrated network seems to be an ideal solution as it is able to provide easy deployment, high speed data rate and wide range coverage with high throughput, low end to end delay, flat and low jitter.Wi-Fi/ WiMAX integrated network provides Quality of Service (QoS) that can support all kinds of real-time application in wireless networks that includes priority scheduling and queuing for bandwidth allocation that is based on traffic scheduling algorithms within wireless networks.In this paper, we have designed a Wi-Fi/ WiMAX integrated network and analyze the performance of different scheduling algorithms for that integrated network and highlight our findings on the scheduling algorithm which will give the best performance for a heterogeneous network.

Motivated by the special characteristics of multimedia tasks, we consider non-preemptive scheduling of tasks where there exists no (or very limited) information concerning the tasks before they are released. We present impossibility results and analyze algorithms for non-preemptive scheduling in single processor and multiprocessor systems. In particular, one set of our results considers the competitive ratio of the scheduling algorithm when the length of the tasks is not greater than C/sub max/ (and not smaller than C/sub min/). We show that the performance of a scheduling algorithm is improved dramatically when the release time of the tasks is O(C/sub max/) prior to their deadline; achieving a competitive ratio that is close to one.

This study focuses on managing equity-conscious radio resources in OFDMA-based WIMAX networks with comparison of three algorithms, Max Rate, RR and PFS. Hence WiMAX Forum has been formed, which promotes the technology and provides compatibility and teroperability for 802.16-based products. The problem in heterogeneous WiMAX, users may possess different transmit power capabilities and experience independent channel realizations. As a result, those with lower power Capabilities or poor channel conditions will starve throughput share. It is observed that users with better channel conditions always dominate usage of the spectrum, causing low throughput for users with poor channel conditions. This paper studied the performance of downlink scheduling algorithms that perform radio resource allocation in order to improve the performance of the WiMAX system. These algorithms are Max Rate, Round Robin, and Proportional Fair. In addition, scheduling algorithm was implementing in order to explain the fairness concept. The algorithms were comparing to each other in terms of system throughput and fairness index. The performance evaluations are carrying out using MATLAB software. Simultaneous results show that the most suitable scheduling algorithm in terms of fairness index and system throughput for the WiMAX network is proportional fair. Meanwhile, Max Rate provides the highest throughput, even though the number of users is increased.

In typical optical burst switching (OBS) networks a reservation request is either accepted or rejected as soon as it is presented to the employed wavelength scheduling algorithm. This paper proposes a class of wavelength scheduling algorithms that process a batch of data bursts together instead of processing them one by one. When a control burst with a reservation request arrives to a free batch scheduler, the scheduler waits for a small amount of time, called the acceptance delay, before deciding to accept or reject the reservation request. After the acceptance delay has passed, the scheduler processes all the reservation requests that have arrived during the acceptance delay period, then it accepts the requests that maximizes the utilization of the wavelength channels. We describe an optimal batch scheduler that serves as an upper bound on the performance of batch scheduling algorithms. Furthermore, we introduce two novel heuristic batch scheduling algorithms. Simulation results suggest that batch schedulers could decrease the blocking probability by 25% compared to the best previously known wavelength scheduling algorithm