Strategic applications for the country, such as the oil and gas industry, meteorology, and bioinformatics, require High-Performance Computing (HPC) for the intensive processing of large volumes of data.
Cloud Computing have been widely used as an excellent low-cost alternative to HPC and offer a vast set of virtualized resources that can be rapidly provisioned and dynamically scaled.
We develop at the C+HPC Laboratory of the Universidade Federal Fluminense (UFF) scientific research aimed at resource allocation and management in computational clouds for the efficient execution of HPC applications, aiming at reducing the associated monetary costs. In this way, we seek to overcome the challenges imposed by the use of this computational paradigm, such as the minimization of time and energy consumption associated with the execution of applications, the implementation of fault tolerance techniques that guarantee the completion of executions, and the observance of agreements existing service level.