Storm surge simulation and load balancing in azure cloud. Chakraborty, A., Pathirage, M., Suriarachchi, I., Chandrasekar, K., Mattocks, C., & Plale, B. In Simulation Series, volume 45, 2013. abstract bibtex Cloud computing platforms are drawing increasing attention of the scientific research communities. By providing a framework to lease computation resources, cloud computing enables the scientists to carry out large-scale experiments in a cost-effective fashion without incurring high setup and maintenance costs of a large compute system. In this paper, we study the implementation and scalability issues in deploying a particular class of computational science applications. Using Platform-as-a-Service (PAAS) of Windows Azure cloud, we implement a high-throughput Storm-Surge Simulation in both a middleware framework for deploying jobs (in cloud and grid environment) and a MapReduce framework - a data parallel programming model for processing large data sets. We present the detailed techniques to balance the simulation loads while parallelizing the application across a large number of nodes.
@inproceedings{
title = {Storm surge simulation and load balancing in azure cloud},
type = {inproceedings},
year = {2013},
volume = {45},
issue = {6},
id = {042339ca-1ac1-32a0-a1e3-8bfef1f36491},
created = {2019-10-01T17:20:45.227Z},
file_attached = {false},
profile_id = {42d295c0-0737-38d6-8b43-508cab6ea85d},
last_modified = {2019-10-01T17:23:16.244Z},
read = {false},
starred = {false},
authored = {true},
confirmed = {true},
hidden = {false},
citation_key = {Chakraborty2013},
folder_uuids = {73f994b4-a3be-4035-a6dd-3802077ce863},
private_publication = {false},
abstract = {Cloud computing platforms are drawing increasing attention of the scientific research communities. By providing a framework to lease computation resources, cloud computing enables the scientists to carry out large-scale experiments in a cost-effective fashion without incurring high setup and maintenance costs of a large compute system. In this paper, we study the implementation and scalability issues in deploying a particular class of computational science applications. Using Platform-as-a-Service (PAAS) of Windows Azure cloud, we implement a high-throughput Storm-Surge Simulation in both a middleware framework for deploying jobs (in cloud and grid environment) and a MapReduce framework - a data parallel programming model for processing large data sets. We present the detailed techniques to balance the simulation loads while parallelizing the application across a large number of nodes.},
bibtype = {inproceedings},
author = {Chakraborty, A. and Pathirage, M. and Suriarachchi, I. and Chandrasekar, K. and Mattocks, C. and Plale, B.},
booktitle = {Simulation Series}
}