Generalized Communicators in the Message Passing Interface
IEEE Transactions on Parallel and Distributed Systems
SOSP '03 Proceedings of the nineteenth ACM symposium on Operating systems principles
Proceedings of the 28th international conference on Software engineering
MapReduce: simplified data processing on large clusters
Communications of the ACM - 50th anniversary issue: 1958 - 2008
On Building Scientific Workflow Systems for Data Management in the Cloud
ESCIENCE '08 Proceedings of the 2008 Fourth IEEE International Conference on eScience
Experimental study of large-scale computing on virtualized resources
VTDC '09 Proceedings of the 3rd international workshop on Virtualization technologies in distributed computing
Hive: a warehousing solution over a map-reduce framework
Proceedings of the VLDB Endowment
CloudWF: A Computational Workflow System for Clouds Based on Hadoop
CloudCom '09 Proceedings of the 1st International Conference on Cloud Computing
Hadoop: The Definitive Guide
Overview of sciDB: large scale array storage, processing and analysis
Proceedings of the 2010 ACM SIGMOD International Conference on Management of data
Hi-index | 0.00 |
The climate research community is increasingly interested in utilizing direct, observational measurements to validate model output in an effort to tune those models to better approximate our planet's dynamic climate. The current emphasis on performing these comparisons at regional, as opposed to global, scales presents challenges both scientific and technical, since regional ecosystems are highly heterogeneous and the available data is not readily consumed on a regional basis. If provided with a common approach for efficiently accessing and utilizing the existing observational datasets, climate researchers have the potential to effect lasting societal, economic and political benefits. A key challenge, however, is that model-to-observational comparison requires massive quantities of data and significant computational capabilities. Further complicating matters is the fact that, currently, observational data and model outputs exist in a variety of data formats, utilize varying degrees of specificity and resolution, and reside in disparate, highly heterogeneous data systems. In this paper we present a software architectural approach that leverages the advantages of cloud computing and modern open-source software technologies to address the regional climate modeling problem. Our system, dubbed RCMES, is highly scalable and elastic, allows for both local and distributed management of the satellite observations and generated model outputs, and delivers this information to climate researchers in a way that is easily integrated into existing climate simulations and statistical tools.