Big data is being used to better understand how diseases affect human populations. Archimedes, an organization that crafts mathematical models of human diseases and interventions, is using the open-source Hadoop platform and Univa's Grid Engine distributed resource-management platform to power its mathematical modeling software.
Univa is a provider of automation and management software for computational and big data infrastructures. Its software allows researchers to map the human genome and is also used by engineers in the automotive and aerospace industries to design vehicles. Companies use Grid Engine to build large computer infrastructures that can process large volumes of data.
Univa announced its latest collaboration with Archimedes on Nov. 27. The data-modeling company has also used Univa's Grid MP distributed computing software.
Archimedes has studies the effects of diseases, tests and treatments since 1993.
The main goal of Archimedes' work is to use big data analytics to help doctors make better decisions, Katrina Montinola, the company's vice president of engineering, told eWEEK. From the information gained from data modeling, patients may be able to make health decisions such as losing weight or quitting smoking.
Researchers use the Archimedes Model to study diseases, health care systems and patient populations as well as customize models for tests and treatments, according to the company. Pharmaceutical companies and government agencies also use the Archimedes Model to answer complex health care questions.
Archimedes' data-modeling tools allow researchers to make connections between health events, said Montinola. Its software can compare a patient's electronic health record with other risk factors to help physicians achieve better health outcomes for patients.
Archimedes used Grid Engine to build an application called Aggregator, which can form correlations between health events in near real-time.
Processing the data has been difficult since it has taken Archimedes' platform a long time to load data sets in the past, according to Montinola. With Grid Engine, researchers and health care providers are able to use the Archimedes data-modeling tools to perform data analysis faster to speed up the delivery of care.
Aggregator, which incorporates Hadoop, the open source big data platform, is part of the Archimedes Healthcare Simulator (ARCHeS) platform, a software as a service (SAAS) that allows researchers to run simulations that specify populations and eligibility criteria as well as the types of interventions and care processes. Researchers can look up information on conditions, treatments and outcomes for patients that may be at risk for a condition such as stroke or diabetes.
Grid Engine allows Archimedes to run ARCHeS and Aggregator on the same cluster without requiring additional hardware. Using Grid Engine cut Archimedes' hardware costs in half and reduced the time needed to process large data sets and perform calculations, said Montinola.
"Archimedes is showing the medical world how to use big data to solve real medical problems, both at an individual and national level by comparing a person's electronic health record with various risk factors," Gary Tyreman, Univa's CEO, said in a statement. "Univa Grid Engine is enabling organizations to unleash the power of Hadoop framework without the heavy costs associated with buying a new cluster to cater to its needs."
By using Grid Engine, Archimedes can also apply scheduling policies to all computing jobs, which makes migration cleaner, said Montinola.