A Self-Training Algorithm for Load Balancing in Cluster Computing

Mohammadpour, P.; Sharifi, Mohsen; Paikan, Ali

doi:10.1109/ncm.2008.178

Cited by 9 publications

(9 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In periodic strategies, individual processors will report their load information to each other periodically at a predetermined time interval. This implies that information may be exchanged, although it might be unnecessary [31]. One of the most critical aspects of periodic policies lies in setting the interval for information exchange.…”

Section: Load Information Exchangementioning

confidence: 99%

A Dynamic Load Balancing Framework for Real-time Applications in Message Passing Systems

Kabbany

Wanas²,

Hegazi

et al. 2010

Int J Parallel Prog

View full text Add to dashboard Cite

Load balancing algorithms are designed essentially to equally distribute the load on processors and maximize their utilities while minimizing the total task execution time. In order to achieve these goals, the load-balancing mechanism should be "fair" in distributing the load across the different processors. This implies that the difference between the heaviest-loaded and the lightest-loaded processors should be minimized. Therefore, the load information on each processor must be updated such that the load-balancing mechanism can be more effective. In this work, we present an application independent dynamic algorithm for scheduling tasks and loadbalancing in message passing systems. We propose a DAG-based Dynamic Load Balancing algorithm for Real time applications (DAG-DLBR) that is designed to work dynamically to cope with possible changes in the load that might occur during runtime. This algorithm addresses the challenge of devising a load balancing scheme which judicially deals with the hybrid execution of existing real-time application (represented by a Direct Acyclic Graph (DAG)) together with newly arriving jobs. The main objective of this algorithm is to reduce response times of the newly arriving jobs while maintaining the time constrains of the existing DAG. To evaluate the performance of the DAG-DLBR algorithm, a comparison with the performance of two common dynamic load balancing algorithms is presented. This comparison is performed by evaluating, experimentally, the execution time of different load balancing algorithms on a homogenous real parallel machine. In addition, the values of load imbalance, the execution time, and the communication overhead time are evaluated analytically using different benchmarks as test-bed workloads. These workloads cover a wide range of dynamic applications with different task types. Experimental results illustrate the improved performance of the DAG-DLBR algorithm compared to both distributed and hierarchal based algorithms by at least 12 and 19%, respectively. This improvement is true for all workloads, even with highly dependent workload. The DAG-DLBR algorithm achieves lower computation time than its corresponding values of both the distributed and the hierarchical-based algorithms for 4, 8, 12 and 16 processors.

show abstract

Section: Load Information Exchangementioning

confidence: 99%

A Dynamic Load Balancing Framework for Real-time Applications in Message Passing Systems

Kabbany

Wanas²,

Hegazi

et al. 2010

Int J Parallel Prog

View full text Add to dashboard Cite

show abstract

“…The proposed resource allocation algorithm is on the basis of FIRSTAVAILABLE and overhead model (5). When job execution time (T) achieves the minimum, Equation (5) can obtain optimal node number (N).…”

Section: A An Overhead Modelmentioning

confidence: 99%

“…The max node number in node list is equal to requested task number. In second step calculate the optimal node number (N) based on (5). And then distribute tasks to available nodes.…”

Section: A An Overhead Modelmentioning

confidence: 99%

An integrated resource allocation algorithm for CPU intensive and I/O intensive jobs in multi-core cluster

Liang

Liu

Xiao

et al. 2010

2010 2nd International Conference on Future Computer and Communication

View full text Add to dashboard Cite

In cluster jobs have different characteristics, such as CPU intensive and I/O intensive. Many resource allocation strategies deal with CPU intensive jobs or I/O intensive jobs.However, few algorithms can deal with both of them. In this paper we propose an integrated resource allocation algorithm, which considers job characteristics. The proposed algorithm is based on an overhead model. The model can integrate the different job characteristics. Simulation results indicate that the proposed resource algorithm can effectively achieve improvement in average execution time compared with FIRSTAVAILABLE. Moreover, proposed algorithm can promote system load balancing to some extent.

show abstract

“…This lead to the deployment of commodity distributed systems. One such system is a high performance cluster computing system which are aimed to create a processing model with single system image [20]. Compute-intensive problems are solved by dividing the given problems into executable tasks [20] which could be processed on a single cluster node.…”

Section: Introductionmentioning

confidence: 99%

“…One such system is a high performance cluster computing system which are aimed to create a processing model with single system image [20]. Compute-intensive problems are solved by dividing the given problems into executable tasks [20] which could be processed on a single cluster node. If an appropriate node is not assigned to this process, a user may conclude up with ending the process in the current node and redistributing the process on a different node, consequently reducing the performance with an increase in the response time.…”

Section: Introductionmentioning

confidence: 99%

Load Balancing for Effective Resource Provisioning in a Heterogeneous Cluster using Machine Learning

Velayutham¹,

Chandrasekaran²

2019

IJEAT

View full text Add to dashboard Cite

Compute Clusters are typically installed to increase performance and/or accessibility. Appropriate Resource Provisioning is a key feature in clustered computing environments to avoid provisioning resources lower than the actual requirement and provisioning of resources in excess. In this paper, a load balancing scheme leading to effective provisioning of resources have been proposed. Job History of compute-intensive jobs have been collected by conducting experiments to observe basic parameters of a job in a heterogeneous computing cluster environment. A Machine Learning model using Multi-Layer Perceptron and Support Vector Machine for provisioning of resources has been presented. The prediction model uses the job history collected from the cluster environment to predict the resource that would be appropriate for provisioning in future. The accuracy of the model is computed and the results of experiments show that Multi-Layer Perceptron presents a better performance than Support Vector Machine

show abstract

A Self-Training Algorithm for Load Balancing in Cluster Computing

Cited by 9 publications

References 6 publications

A Dynamic Load Balancing Framework for Real-time Applications in Message Passing Systems

A Dynamic Load Balancing Framework for Real-time Applications in Message Passing Systems

An integrated resource allocation algorithm for CPU intensive and I/O intensive jobs in multi-core cluster

Load Balancing for Effective Resource Provisioning in a Heterogeneous Cluster using Machine Learning

Contact Info

Product

Resources

About