Currently we have the following applications:


tau-loptau-Lop is a new parallel performance model aimed to help in the design and optimization of parallel algorithms inside multicore clusters. It represents a parallel algorithm and predicts accurately its costs through the concept of concurrent transfers.

By now, tau-Lop has been applied to underlying algorithms in MPICH and Open MPI mainstream implementations of some MPI collectives in shared memory. Current work is in the application of the model to collective operations when deployed in networks of multicore nodes.

Read More

Iberian Ham

JamónIberian Ham: We have developed a real case application: A software application containing three modules was used for the analysis of MRI (Magnetic Resonance Imaging).

The initial module aimed to detect the Biceps femoris muscle by using Active Contours.

The second module consisted in the selection procedure for the Region of Interest (ROI) on each image; this selection drew up the maximum rectangular area on the muscle.

The third and last module included the analysis of the ROIs by applying three common methods in computational texture analysis, which require the use of rectangular images. All three methods used matrices based on second order statistics. The first one, GLCM (Grey Level Coocurrence Matrix), was constructed with information of the complete ROI. The second one, the so-called NGLDM (Neighbouring Grey Level Dependence Matrix), gathered information from square neighbourhoods inside the ROI. The third one, the GLRLM (Grey Level Run Length Matrix), only accounted for information about lineal segments of the ROI.

Read more

Log in