Conciencia Sísmica (10/11/2017 03:32 PM)
- Visualization Architecture For Scientific Applications As A Service (01/28/2017 04:50 PM)
Support Vector Machines with MPP platform (07/16/2016 01:57 PM)
The development of different branches of science and its impact, create new challenges and needs to be solved by engineering. This is the case of research, chemistry, astronomy, biology, among others, which generate thousands of data in short time intervals, and in many cases requires classification decisions.Although current architectures are efficient, some machine learning algorithms for classification takes time considerably high. In this paper the problem is attacked by implementing parallel programming standard OpenACC, OpenMP and MPI on LIBSVM, which is a library that implements algorithms for support vector machines. The product of this work not only ensures a massive data processing time considerably low, but also fidelity in the results obtained by this...
Processing and Visualization in Embedded Architectures of High Performance Computing (07/15/2016 02:44 PM)
- Propose an Embedded Architecture of High Performance Computing for image processing and visualization.
- Identify the components and relationships of the system "microscope - computer unit (SBC)", to propose an embedded architecture of processing and visualization of high performance....
enerGyPU for Monitoring Performance and Power Consumption on Multi-GPUs (11/10/2015 11:24 AM)
enerGyPU is a tool that can be used to analyze multiple tests under different combinations of parameters to observe the key factors that determine the energy efficiency in terms of ‘Energy per Computation’ on Cluster with Multi-GPUs. enerGyPU is a monitor to centralize and automate the data captured in runtime while is executed in parallel with the Linpack Benchmark and displays information via sequenceplots, statistical tables, bar graph and shows results in terms of energy efficiency....