Dynamic Load Balancing and Job Replication in a Global-Scale Grid Environment: A Comparison

Dobber, Menno; Mei, Rob van der; Koole, Ger

doi:10.1109/tpds.2008.61

Cited by 40 publications

(20 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The topic of replication of tasks has been extensively explored in the context of Grid systems [16], [17], [18], [19], [20] without addressing the issue of cost for resource utilization. Plankensteiner and Prodan [21] proposed an algorithm for scheduling of workflow applications on Grids with tasks replication to meet deadline constraints.…”

Section: Related Workmentioning

confidence: 99%

“…1: Data: V M: List of provisioned VMs. 2: for each virtual machine vm in V M do 3: t start ← first task scheduled to vm; 4: t last ← last task scheduled to vm; 5: vm.startTime ← st(t start ); 6: transferTime ← 0; for each parent t p of t start do 8: if D(e p,start ) > transferTime then 9: transferTime ← D(e p,start ); 10: end if 11: end for 12: vm.startTime ← vm.startTime − transferTime − bootTime; 13: for each child t c of t end do 14: if D(e end,c ) > transferTime then 15: transferTime ← D(e end,c ); 16: end if 17: end for 18: vm.endTime ← vm.endTime + transferTime; 19 vm ← first VM from V whose cost is smaller than rb; 10: if no VM was found then 11: Break; 12: end if 13: Replicate vm; 14: Add the new machine to V M; 15: Add a slot representing the new machine in P ; 16: Move vm to the end of V ; 17: rb ← cost of vm; 18: end while 19: Sort P and U in increasing order of size; 20: Add U at the end of P ; 21: Sort tasks from T in ascending order of replication precedence; 22: for each slot s in P do 23: if s is unpaid and there is no available budget then 24: Break; 25 …”

Section: Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Meeting Deadlines of Scientific Workflows in Public Clouds with Tasks Replication

Calheiros

Buyya

2014

IEEE Trans. Parallel Distrib. Syst.

202

118

View full text Add to dashboard Cite

Abstract-The elasticity of Cloud infrastructures makes them a suitable platform for execution of deadline-constrained workflow applications, because resources available to the application can be dynamically increased to enable application speedup. Existing research in execution of scientific workflows in Clouds either try to minimize the workflow execution time ignoring deadlines and budgets or focus on the minimization of cost while trying to meet the application deadline. However, they implement limited contingency strategies to correct delays caused by underestimation of tasks execution time or fluctuations in the delivered performance of leased public Cloud resources. To mitigate effects of performance variation of resources on soft deadlines of workflow applications, we propose an algorithm that uses idle time of provisioned resources and budget surplus to replicate tasks. Simulation experiments with four well-known scientific workflows show that the proposed algorithm increases the likelihood of deadlines being met and reduces the total execution time of applications as the budget available for replication increases.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Algorithmmentioning

confidence: 99%

Meeting Deadlines of Scientific Workflows in Public Clouds with Tasks Replication

Calheiros

Buyya

2014

IEEE Trans. Parallel Distrib. Syst.

202

118

View full text Add to dashboard Cite

show abstract

“…The multi-tier hierarchical cloud system architecture [16][17][22][23][24][25] supports an efficient method for sharing data and computational and other resources, as shown in Fig.2. It typically consists of different tiers of data centers with different regions and sizes.…”

Section: System Modelmentioning

confidence: 99%

Modeling a Dynamic Data Replication Strategy to Increase System Availability in Cloud Computing Environments

Sun

Chang

Gao

et al. 2012

J. Comput. Sci. Technol.

126

View full text Add to dashboard Cite

Failures are normal rather than exceptional in the cloud computing environments. To improve system availability, replicating the popular data to multiple suitable locations is an advisable choice, as users can access the data from a nearby site. This is, however, not the case for replicas which must have a fixed number of copies on several locations. How to decide a reasonable number and right locations for replicas has become a challenge in the cloud computing. In this paper, a dynamic data replication strategy is put forward with a brief survey of replication strategy suitable for distributed computing environments. It includes: 1) analyzing and modeling the relationship between system availability and the number of replicas; 2) evaluating and identifying the popular data and triggering a replication operation when the popularity data passes a dynamic threshold; 3) calculating a suitable number of copies to meet a reasonable system byte effective rate requirement and placing replicas among data nodes in a balanced way; 4) designing the dynamic data replication algorithm in a cloud. Experimental results demonstrate the efficiency and effectiveness of the improved system brought by the proposed strategy in a cloud.

show abstract

“…Since behavior of node in the system is not static due to node failure or node removal etc., therefore a temporal unbalance among the nodes exists. Due to dynamic nature of grid system a job replication approach [5] are suitable to deal with job processing. A job replication scheme is same as job distribution but send a copy of job to each processor for processing instead of different types of jobs.…”

Section: Disscussionmentioning

confidence: 99%

Load Balancing Approaches in Grid Computing Environment

Pandey¹,

Verma²,

Tamta³

2013

IJCA

View full text Add to dashboard Cite

Grid computing is a kind of distributed computing that involve the integrated and collaborative use of distributed resources. It involves huge amounts of computational task which require reliable resource sharing across computing domains. Load balancing in grid is a technique which distributes the workloads across multiple computing nodes to get optimal resource utilization, minimum time delay, maximize throughput and avoid overload. It is a challenging problem that has been studied extensively is the past several years. This paper attempts to provide a comprehensive overview of load balancing in grid computing environment and also analyses the job distribution and system behavior. Furthermore, this survey various load balancing algorithms for the grid computing environment, identify several comparison metrics for the load balancing algorithms and carry out the comparison based on these identified metrics between them. it also reviews the latest research activities in the area of grid computing, including characteristics, capabilities, architecture, applications, design constraints, scheduling and load balancing and presents a set of challenges and problems.

show abstract

Dynamic Load Balancing and Job Replication in a Global-Scale Grid Environment: A Comparison

Cited by 40 publications

References 16 publications

Meeting Deadlines of Scientific Workflows in Public Clouds with Tasks Replication

Meeting Deadlines of Scientific Workflows in Public Clouds with Tasks Replication

Modeling a Dynamic Data Replication Strategy to Increase System Availability in Cloud Computing Environments

Load Balancing Approaches in Grid Computing Environment

Contact Info

Product

Resources

About