Near-optimal dynamic task scheduling of independent coarse-grained tasks onto a computational grid

Fujimoto, Noriyuki; Hagihara, Kenichi

doi:10.1109/icpp.2003.1240603

Cited by 69 publications

(40 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There has been much work performed on load balancing independent tasks (bag of tasks) in the distributed and heterogeneous computing fields [29,30,31,32]. Many of the scheduling algorithms try to minimize the makespan, which can be considered a type of load balancing.…”

Section: Related Workmentioning

confidence: 99%

Dynamic load balancing for petascale quantum Monte Carlo applications: The Alias method

Sudheer

Krishnan

Srinivasan

et al. 2013

Computer Physics Communications

View full text Add to dashboard Cite

Diffusion Monte Carlo is a highly accurate Quantum Monte Carlo method for the electronic structure of materials, but it requires frequent load balancing or population redistribution steps to maintain efficiency on parallel machines. This step can be a significant factor affecting performance, and will become more important as the number of processing elements increases. We propose a new dynamic load balancing algorithm, the Alias Method, and evaluate it theoretically and empirically. An important feature of the new algorithm is that the load can be perfectly balanced with each process receiving at most one message. It is also optimal in the maximum size of messages received by any process. We also optimize its implementation to reduce network contention, a process facilitated by the low messaging requirement of the algorithm: a simple renumbering of the MPI ranks based on proximity and a space filling curve significantly improves the MPI Allgather performance. Empirical results on the petaflop Cray XT Jaguar supercomputer at ORNL showing up to 30% improvement in performance on 120,000 cores. The load balancing algorithm may be straightforwardly implemented in existing codes. The algorithm may also be employed by any method with many near identical computational tasks that requires load balancing.

show abstract

Section: Related Workmentioning

confidence: 99%

Dynamic load balancing for petascale quantum Monte Carlo applications: The Alias method

Sudheer

Krishnan

Srinivasan

et al. 2013

Computer Physics Communications

View full text Add to dashboard Cite

show abstract

“…This type of scheduling works well and at non homogenous environments such as grid is suitable but because of calculations and communications of overload the efficiency of system is lower than optimized position. Other scheduling strategies: There are many classic strategies that are used in calculation grids such as PUNCH,Optimal Job Scheduling, Using Legacy Codes, [5] In all of these strategies are determined the number of sending tasks to any processing source by scheduled module. The variations which are accrued cant effect on scheduled, really there isn't any feedback and these strategies cant scheduled without interruption.…”

Section: The Initial Algorithm and Rqsg-imentioning

confidence: 99%

“…[4] In the past task scheduling was proposed for super computers and parallel computers . [5][6] [7][8]…”

Section: The Initial Algorithm and Rqsg-imentioning

confidence: 99%

A Scheduling Algorithm for Increasing the Quality of the Distributed Systems by using Genetic Algorithm

Delavar¹,

Aghazarian²,

Litkouhi³

et al. 2011

IJIET

View full text Add to dashboard Cite

Abstract-This article will present a scheduling algorithm for increasing the quality of the distributed system by using genetic algorithm. The proposed algorithm OQSG is more efficient in comparison with the previous algorithms. This proposed algorithm can be reliable like the other algorithms, regarding the effective parameters used for the available scheduling. OQSG algorithm can create more efficient parameters by using the priority function that is more reliable by using the scheduling the most used works. The proposed algorithm can use the least possible time in comparison with the previous algorithm, using new techniques in the purpose function. This condition helps us to use the effective parameters, which increase the system efficiency, after several assessment used in the heterogeneous purpose function. The proposed algorithm can have the optimum selection in processing and allocating the resources, considering maximum and minimum scheduling of works. This work is fulfilled when we decrease the complexity of time. Finally, we could create the efficiency of distributed systems with high quality according to different simulation results.

show abstract

“…Most of them address divisible load scheduling in Grids, see, for instance, Robertazzi and Yu [9]. Fujimoto and Hagihara [3] discuss the scheduling of sequential independent jobs on systems with processor speeds that vary over time and between different machines. They claim that the makespan objective is not applicable and propose a different criterion based on total processor cycle consumption.…”

Section: Introductionmentioning

confidence: 99%

Online scheduling in grids

Schwiegelshohn

Tchernykh

Yahyapour

2008

2008 IEEE International Symposium on Parallel and Distributed Processing

View full text Add to dashboard Cite

show abstract

Near-optimal dynamic task scheduling of independent coarse-grained tasks onto a computational grid

Cited by 69 publications

References 13 publications

Dynamic load balancing for petascale quantum Monte Carlo applications: The Alias method

Dynamic load balancing for petascale quantum Monte Carlo applications: The Alias method

A Scheduling Algorithm for Increasing the Quality of the Distributed Systems by using Genetic Algorithm

Online scheduling in grids

Contact Info

Product

Resources

About