Optical burst switching (OBS) is an emerging technology that allows variable size data bursts to be transported directly over DWDM links. In order to make OBS a viable solution, the wavelength scheduling algorithms ne...
详细信息
Optical burst switching (OBS) is an emerging technology that allows variable size data bursts to be transported directly over DWDM links. In order to make OBS a viable solution, the wavelength scheduling algorithms need to be able to utilize the available wavelengths efficiently, while being able to operate fast enough to keep up with the burst incoming rate. For example, for a system with 32 wavelengths per link, each operating at 10 Gb/s, we need to process one BHC every 25 ns, in order to support an average burst length of 1 KByte. When implemented in hardware, the running time of the well-known horizon scheduler is O(1) for practical numbers of wavelengths. Unfortunately, horizon scheduling cannot utilize the voids created by previously scheduled bursts, resulting in low bandwidth utilization. To date, Min-SV is the fastest scheduling algorithm that can schedule wavelengths efficiently. However, its complexity is O(log m) and it requires lOlog m memory accesses to schedule a single burst. This means that it can take up to several microseconds for each burst request, which is still too slow to make it a practical solution for OBS deployment. In this paper, we propose an optimal wavelength scheduler using constant time burst resequencing (CTBR), which runs in O(1) time. The proposed CTBR scheduler is able to produce optimal wavelength schedules while having the same processing speed as the horizon scheduler. The algorithm is well-suited to high performance hardware implementation.
A bandwidth-efficient video streaming algorithm over the cdma2000 1x EV-DO network using GOP-based priority scheduling is proposed in this work. The proposed algorithm performs an application adaptation by considering...
详细信息
A bandwidth-efficient video streaming algorithm over the cdma2000 1x EV-DO network using GOP-based priority scheduling is proposed in this work. The proposed algorithm performs an application adaptation by considering the postchannel status to cope with the fluctuating wireless mobile radio channel. The peak-signal-to-noise-ratio (PSNR) value and the number of buffer underflows at the mobile terminal (MT) are used as performance metrics to indicate the spatial and temporal quality of received video, respectively. It is demonstrated by computer simulation that the proposed priority scheduling algorithm adapts source bit rates of pre-encoded video streams to the bandwidth fluctuation of wireless channels well and yields satisfactory spatial and temporal video quality.
A resource allocation policy using transmit Beam- forming is proposed for the downlink of multiuser systems. Users to be selected for transmission in a given time slot are examined from groups of users to which the pr...
详细信息
A resource allocation policy using transmit Beam- forming is proposed for the downlink of multiuser systems. Users to be selected for transmission in a given time slot are examined from groups of users to which the proposed algorithm determines the resources to be allocated in terms of antennas and power. The transmit beam-vectors are computed according to a zero-forcing criterion. Users are selected from different groups so as to achieve a desired balance between maximizing throughput and achieving a high degree of fairness among groups of users as well as between users within a group. Such grouping allows the proposed algorithm to make full usage of the resources and to dynamically allocate these while avoiding the complexity of the NP-complete resource reservation and users' scheduling. Computational complexity is reduced by first estimating the resources to be allocated to the groups of users and then distributing these fairly among the selected users in each group. Such an approach allows the base station to assign its antenna and power resources to groups of users for which transmission would alleviate the interference that would result in neighboring cells, or to direct the resources toward overloaded regions in the cell. Simulation results and comparisons show that the proposed dynamic resource allocation strategy achieves multiuser diversity and a high efficiency in terms controlling the tradeoff between throughput and fairness.
Consider the problem of scheduling a set of periodically arriving tasks on a multiprocessor with the goal of meeting deadlines. Processors are identical and have the same speed. Tasks can be preempted and they can mig...
详细信息
Consider the problem of scheduling a set of periodically arriving tasks on a multiprocessor with the goal of meeting deadlines. Processors are identical and have the same speed. Tasks can be preempted and they can migrate between processors. We propose an algorithm with a utilization bound of 66% and with few preemptions. It can trade a higher utilization bound for more preemptions and in doing so it has a utilization bound of 100%
Earlier work has developed the underpinnings of a theory of scheduling computations having intertask dependencies - modeled via dags - for Internet-based computing. The goal of the schedules produced is to render task...
详细信息
Earlier work has developed the underpinnings of a theory of scheduling computations having intertask dependencies - modeled via dags - for Internet-based computing. The goal of the schedules produced is to render tasks eligible for execution at the maximum possible rate. This goal aims: (a) to utilize remote clients’ computational resources well, by always having work to allocate to an available client; (b) to lessen the likelihood of the "gridlock" that ensues when a computation stalls for lack of eligible tasks. The dags handled by the theory thus far are those that can be constructed from a given collection of bipartite building-block dags via the operation of dagcomposition. The current paper extends the range of applicability of the theory by significantly expanding the repertoire of building-block dags that the scheduling algorithms can handle. Thereby, the theory can now schedule large classes of "expansive" and "reductive" dags optimally.
Grids have emerged as paradigms for the next generation parallel and distributed computing. Computational grid can be defined as large-scale high-performance distributed computing environments that provide access to h...
详细信息
Grids have emerged as paradigms for the next generation parallel and distributed computing. Computational grid can be defined as large-scale high-performance distributed computing environments that provide access to high-end computational resources. Grid scheduling is the process of scheduling jobs over grid resources. Improving overall system performance with a lower turn around time is an important objective of grid scheduling. In this paper a priority based scheduling algorithm is proposed. In this algorithm a new parameter named "priority" has been taken into consideration. The algorithm classifies the jobs into high, medium and low categories based on their priority. The priority assignment is done by considering the computational power of job and level of parallelism. The value for level of parallelism is assigned based on the amount of parallelism exhibited by the job and the amount of parallelism exhibited by the available resources. Generally, a job, which needs high computational power and exhibits low parallelism is given a high priority. Prioritizing the jobs in this way can improve the performance of computational grids. The effectiveness of our algorithm is evaluated through simulation results and its superiority over other known algorithms is demonstrated
In the paper, novel sessionless approach to test-schedulling is presented. It utilizes so-called STEPs during special random-search based scheduling algorithm. The algorithm explores the state-space of so-called i-sch...
详细信息
In the paper, novel sessionless approach to test-schedulling is presented. It utilizes so-called STEPs during special random-search based scheduling algorithm. The algorithm explores the state-space of so-called i-schedules whereas an i-schedule is an integer-vector encoded test-schedule represented by n-touple of STEPs. Proposed algorithm tries to link tests to STEPs in such a way there are no resource sharing conflicts in the best-found test schedule and hopefully, test schedule constraints are met maximally at minimal time and TAM values
This paper investigates real-time scheduling in a system whose energy reservoir is replenished by an environmental power source. The execution of tasks is deemed primarily energy-driven, i.e., a task may only respect ...
详细信息
ISBN:
(纸本)9780769526195;0769526195
This paper investigates real-time scheduling in a system whose energy reservoir is replenished by an environmental power source. The execution of tasks is deemed primarily energy-driven, i.e., a task may only respect its deadline if its energy demand can be satisfied early enough. Hence, a useful scheduling policy should account for properties of the energy source, capacity of the energy storage as well as power dissipation of the single tasks. We show that conventional scheduling algorithms (like e.g. EDF) are not suitable for this scenario. Based on this motivation, we state and prove optimal scheduling algorithms that jointly handle constraints from both energy and time domain. Furthermore, an offline schedulability test for a set of periodic or even bursty tasks is presented. Finally, we validate the proposed theory by means of simulation and compare our algorithms with the classical earliest deadline first algorithm
Concurrent realization method of process planning and production scheduling was suggested in the case of integrated environment, establishes the models of process planning and production scheduling integrated system. ...
详细信息
Concurrent realization method of process planning and production scheduling was suggested in the case of integrated environment, establishes the models of process planning and production scheduling integrated system. The models can classify process planning and production scheduling into different stages. The relations between process planning and production scheduling are analyzed aiming at process planning at the different stages and the tasks of production scheduling. The concurrent operation model is adopted to make the both closely combined with each other, on the basis of which, this paper discusses the effects of the concurrent process plan upon the disciplinarians of scheduling results, and advances the realization of multi-process alternative scheduling algorithm. The simulation tests indicate that using multi-process patterns to carry out scheduling is better than mono-using basic process plan to schedule so that the scheduling complexity is greatly reduced. And the maximum flow time and mean flow time of working pieces is reduced by 30% and 14% respectively, while machine-tool utilization rate is raised by 40%. This indicates that under the integrated patterns, the concurrent integration of process planning and production scheduling is an effective way to improve the efficiency of the production system
In this paper we consider bandwidth-efficient means for the user equipment (UE) to conduct channel quality indication (CQI) for OFDMA frequency domain packet scheduling (FDPS). Using a UTRAN long-term evolution (LTE) ...
详细信息
In this paper we consider bandwidth-efficient means for the user equipment (UE) to conduct channel quality indication (CQI) for OFDMA frequency domain packet scheduling (FDPS). Using a UTRAN long-term evolution (LTE) system model, we find that a threshold-based CQI scheme can enable about 60% CQI bandwidth saving with less than 5-10% loss on system performance compared to near-ideal signaling. To sustain sufficient CQI reporting accuracy and high FDPS gain, we show that time- domain CQI averaging over multiple sub-frames is desirable although it lowers the mobility performance. Conducting the averaging on the UE side is most attractive from a signalling perspective. Combining averaging with lowered CQI reporting rate, FDPS can provide around 30-50% gain for LTE with a CQI reporting bandwidth as low as 10 kbit/s in a 20 MHz system (up to 3-5 km/h).
暂无评论