Genomic research generates large amounts of data requiring complex informatics analysis focused on obtaining significant biological information. This requires high performance computing (HPC) also known as supercomputing.

The Supercomputer and Information Technology Unit at INMEGEN that processes all the data generated on the Laboratories and High Technology Units is equipped with a robust  infrastructure developed by IBM. It is capable of performing 2,200,000,000,000 floating-point operations per second. This Linux cluster consists of 120 nodes or servers and each nodehas two AMD dual-core Opteron processors at 2.0 Ghz with 5GB of ram.

The system also has four special nodes, three of them have  two AMD quad-core processors at 2.6 Ghz with 64GB of ram, and the fourth node has two eight-core power5+ processors with 128GB of ram.

As for storage a fault-tolerant disk array with 15 terabytes storage capacity is set up. This is where the data from the laboratories and Hight Technologty Units goes and is made available to all the nodes in the cluster.

The backup system consist of a tape robot with 120 terabytes storage capacity where the data from the disk array is being stored daily.

Several bioinformatics applications are being run in the cluster. One of the most prominent analysis that ran on it is The Mexican Genome Diversity Project which can be consulten on the following link http://diversity.inmegen.gob.mx.


Responsible: Fernando Robles