Lab Canada

Genomics company to use server clusters to speed up research

Quebec City, QC October 1, 2003 IBM says that population genetics specialist firm Galileo Genomics has chosen IBM servers to run its laboratory, clinical and business applications. This standardization will improve efficiencies for laboratory operations and research efforts.

Galileo has chosen IBM eServer xSeries to run business applications such as e-mail and file sharing, and mission-critical applications such as its Laboratory Instrument Management System, which controls laboratory work flow, robotics and data management.

Additionally, Galileo is piloting a cluster of IBM’s eServer xSeries running Linux to analyze genetic data from the Quebec Founder Population, a homogenous population of six million distantly related Quebec residents, to identify genes that cause common diseases such as diabetes, schizophrenia and asthma. Discoveries will lead to targeted drug therapies that treat the root genetic cause of disease rather than the symptoms, as most drugs do today.

The calculations required to help pinpoint disease genes are computationally intensive and require massive amounts of reliable computing power. Previously, Galileo’s researchers would divide the computational jobs and execute them on dedicated servers, manually combining the results in a final step. Clustering technology now allows Galileo to automatically distribute processing amongst all servers, increasing efficiency and capacity utilization. The Linux cluster is a cost-effective way for Galileo to shorten its processing and discovery time.

“Galileo has grown from a staff of 15 to over 80 employees in two years. This has presented us with various challenges, one of them being an IT infrastructure built on ‘white box’ servers that had reached its capacity to support our discovery process,” says Jean-Franois Lvesque, senior director of IT, Galileo Genomics. “We chose IBM technology for several reasons local support, functionality, pricing and reputation all playing a factor. But the most important reason was our need for robust servers to reliably run applications 24 hours a day with minimal downtime.”