Scalability of grid computing applications

• Grant number: 09/09847-0
• Support Opportunities: Regular Research Grants
• Start date: February 01, 2010
• End date: July 31, 2012
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Hermes Senger
• Grantee: Hermes Senger
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

Computational grids have been widely adopted to provide support for the collaborative work of large communities of scientists and researchers in several fields of science. The grid has been created to provide the aggregation and coordination of significant quantities of multi-institutional, and geographically distributed computational resources with non trivial levels of quality of service. Typically, computational and communication resources in a grid may be heterogeneous, non-dedicated, located at different administrative domains, and subject to periods of failure and unavailability. In such environments, obtaining updated information on the capacity and current status of an abundant amount of resources is a challenging work. In this context, the efficient scheduling of application tasks in not straightforward. The present proposal aims at improving the performance and the scalability of computing intensive applications executing on distributed platforms such as clusters, multi-clusters, and grids. Firstly, the scalability of a class of applications named Bag-of-Tasks will be investigated. Bag-of-Tasks applications are composed of independent tasks which do not communicate to each other and may be executed in any order. Then, a rigorous scalability analysis of such applications running on typical grid platforms such as the master-slave and hierarchical topology will be provided. Finally, strategies and techniques for aiming to improve the scalability of the applications will be proposed and assessed. Furthermore, other application models and platforms may also be evaluated. Finally, novel paradigms for the discovery and allocation of resources will be investigated in the context of scalability of grid applications. (AU)

Scientific publications

(4)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

DA SILVA, FABRICIO A. B.; SENGER, HERMES. Scalability limits of Bag-of-Tasks applications running on hierarchical platforms. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, v. 71, n. 6, p. 14-pg., 2011-06-01. (09/09847-0)

DA SILVA, FABRICIO A. B.; SENGER, HERMES. Scalability limits of Bag-of-Tasks applications running on hierarchical platforms. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, v. 71, n. 6, SI, p. 788-801, JUN 2011. (09/09847-0)

GAIOSO, ROUSSIAN; GIL-COSTA, VERONICA; GUARDIA, HELIO; SENGER, HERMES; MERELLI, I; LIO, P; KOTENKO, I. A Parallel Implementation of WAND on GPUs. 2018 26TH EUROMICRO INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, AND NETWORK-BASED PROCESSING (PDP 2018), v. N/A, p. 8-pg., 2018-01-01. (18/00452-2, 09/09847-0, 17/24500-3)

SENGER, HERMES; BARBOSA DA SILVA, FABRICIO ALVES. BOUNDS ON THE SCALABILITY OF BAG-OF-TASKS APPLICATIONS RUNNING ON MASTER-SLAVE PLATFORMS. PARALLEL PROCESSING LETTERS, v. 22, n. 2, p. 16-pg., 2012-06-01. (09/09847-0)