Job scheduling policy for high throughput computing environments

Abawajy, Jemal

doi:10.1109/icpads.2002.1183466

Cited by 9 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Let w denote the workload of the application scheduled by grid scheduler for this resource. As in [14,24], we assume that the resource has two stage hyper-exponential service time distribution. In addition to workload w, there may be some other jobs (patient and impatient) come from local scheduler (see Fig.…”

Section: Prediction Of Single Resource Performancementioning

confidence: 99%

Fault-aware grid scheduling using performance prediction by workload modeling

Kalantari

Akbari

2008

J Supercomput

View full text Add to dashboard Cite

Computational grids hold great promise in utilizing geographically separated heterogeneous resources to solve large-scale complex problems. However, they suffer from a number of major technical hurdles, including distributed resource management and effective job scheduling. The main focus of this work is devoted on online scheduling of real time applications in distributed environments such as grids. Specifically, we are interested in applications with several independent tasks, each task with a prespecified lifecycle called deadline. Here, our goal is to schedule applications within an optimum overall time considering the specified deadlines. To achieve this, the resource performance prediction based on workload modeling and with the help of queuing techniques is employed. Afterward, a mathematical neural model is used to schedule the subtasks of the application. The main contributions of this work is to incorporate the impatiency factor as well as resource fault in performance modeling of nondedicated distributed systems, and also presenting an efficient and fast parallel scheduling algorithm under time constraint and heterogeneous resources. The proposed model is appropriate for implementation on parallel machines and in O(1) time. The new model was implemented on GridSim toolkit and under various conditions and with different parameters to evaluate the performance of scheduling algorithm. Simulation outcomes have shown that approximately in 87.8% of cases, our model schedules the tasks in such a way that all constraints are satisfied.

show abstract

Section: Prediction Of Single Resource Performancementioning

confidence: 99%

Fault-aware grid scheduling using performance prediction by workload modeling

Kalantari

Akbari

2008

J Supercomput

View full text Add to dashboard Cite

show abstract

“…This is called matchmaking [11]. Now for simulation based approach Condor is used for simulation of the comparison based analysis of two scheduling policies which are Round Robin Opportunistic pol9icy and Multi-level Queue Opportunistic policy in the paper [7]. By effective study of this paper reveals that, for simulation based performance analysis of different type of scheduling policies, Condor is one of the effective way that provides high throughput computing environment.…”

Section: Related Workmentioning

confidence: 99%

“…This file handling is done by condor and Condor is very first to calculate [4][5] and here condor is needed because CPU burst of the processes are very high and we have to have calculate the times of completion of the processes of each queue and the calculating the remaining burst and waiting time all this things. So condor computing environment [7] [9] is a very helpful way to solve this computation problem.…”

Section: P2mentioning

confidence: 99%

Enhancing the Performance of Feedback Scheduling

Bhunia¹

2011

IJCA

View full text Add to dashboard Cite

Feedback scheduling is a kind of process scheduling mechanism where process doesn't come with any priority. According to the CPU burst needed by the process and the CPU burst remaining the processes are shifted between queues of the feedback scheduler to get completed. In multilevel feedback queue the total architecture is divided into multiple prioritised queues. In this paper, we give an approach for jobs which starve in the lower priority queue for long time to get CPU cycle. As a result response time of those starved processes decreases eight to ten percent and over all turn around time of the whole scheduling process decreases around eight to ten percents. In comparison to other types of MLFQs the performance of the proposed scheduling technique is better and practical according to the consequence.

show abstract

“…However, some previous papers did not address them together: they addressed several facets under the limited conditions such as short-term periods, unique service provider, etc [7][8][9][10][11][12][13][14]. In this paper, we will present a fairly-practical model: multiple services, multiple users, and for the delay bound scheduling, we only consider those flows with good channel quality.…”

Section: Introductionmentioning

confidence: 99%

“…Consider also that the channel is a variable. Since there are so many changing variables, we need to investigate how to guarantee the delay bound for every service, fairness for every user and maximum throughput available [6][7]. This process is complicated to solve and is NP-complete.…”

Section: Introductionmentioning

confidence: 99%

A QoS-aware Routing for Multi-Services in Wireless Sensor Networks

Lin¹,

Jiang²

2014

TOAUTOCJ

View full text Add to dashboard Cite

Abstract:In this paper we propose a new routing algorithm for emergency bound based on multiple services for wireless sensor networks. In this routing algorithm channel conditions are associated with each time unit to provide overall fairness and maximize the total throughput of wireless sensor network. The proposed system model supports multiple users of each sensor, each user may have several service demands, every demand has different characteristics such as emergency bound, etc. How to guarantee the fairness for each user and the emergency bound for each service in wireless sensor network is the problem to be solved. In addition, we investigate the channel condition for each user and provide the maximum throughput for the single channel of each time unit.

show abstract

Job scheduling policy for high throughput computing environments

Cited by 9 publications

References 7 publications

Fault-aware grid scheduling using performance prediction by workload modeling

Fault-aware grid scheduling using performance prediction by workload modeling

Enhancing the Performance of Feedback Scheduling

A QoS-aware Routing for Multi-Services in Wireless Sensor Networks

Contact Info

Product

Resources

About