Advanced search
Start date

Scheduling of scientific workflows on clouds

Grant number: 14/08607-4
Support type:Scholarships abroad - Research Internship - Doctorate
Effective date (Start): July 01, 2014
Effective date (End): June 30, 2015
Field of knowledge:Physical Sciences and Mathematics - Computer Science - Computer Systems
Principal Investigator:Edmundo Roberto Mauro Madeira
Grantee:Thiago Augusto Lopes Genez
Supervisor abroad: Rizos Sakellariou
Home Institution: Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
Local de pesquisa : University of Manchester, England  
Associated to the scholarship:12/02778-6 - Dynamic scheduling of multiple workflows for SaaS/PaaS cloud providers considering two SLA levels, BP.DR


The workflow paradigm has become the standard to represent complex scientific (e-Science) problems and their execution flows. With the increasing of the processing demands of these applications upon a massive data sets, more computational demand is being required to execute them. Cloud computing is becoming a financially sustainable computing environment for the owner of a private cloud to run these (or part of) workflow applications. In order to fulfill the application demands, schedulers for such cloud environment have a fundamental role in deciding which types of instances should be leased from a public cloud providers to be joined to the available in-house resources. In the meantime, several applications can be submitted to run at the same time and the scheduler should produce a schedule for each application such that satisfies its demand of quality of service (QoS) requirements. Nevertheless, imprecise information on the input data provided to the scheduler can lead to a misleading schedules and as a result can negatively impact the performance of applications. In this document we propose a detailed study of scheduling algorithms designed to handle workflows on clouds. The aim of this research is the development a dynamic and online algorithm for scheduling multiple workflows concurrently under non-clairvoyant condition, where the available information used as input to the scheduler is uncertain or unknown, e.g., not stationary. Dynamic because the available information of the computational resources are subject to change unpredictably over time, while online due to the workflows can arrive at different time instants and the scheduler should cope with them concurrently. Through algorithms developed in this research, we intend to achieve qualified schedules that ensures the expected performance of the concurrent executions of applications on clouds, regardless of the uncertain information used as input to the scheduler. (AU)

Scientific publications
(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)
GENEZ, THIAGO A. L.; BITTENCOURT, LUIZ F.; DA FONSECA, NELSON L. S.; MADEIRA, EDMUNDO R. M. Estimation of the Available Bandwidth in Inter-Cloud Links for Task Scheduling in Hybrid Clouds. IEEE TRANSACTIONS ON CLOUD COMPUTING, v. 7, n. 1, p. 62-74, JAN-MAR 2019. Web of Science Citations: 2.

Please report errors in scientific publications list by writing to: