Virginia Bioinformatics Institute (VBI) has selected Bright Cluster Manager® software as a new resource for its Network Dynamics and Simulation Science Laboratory (NDSSL), which focuses on building and analyzing virtual cities. VBI will use Bright Cluster Manager for Big Data in a new big data cluster to test new technology to speed up and improve simulations. Testing results will guide VBI’s decisions on larger systems needed for more advanced models focusing on social unrest and infectious disease transmission.
Research at VBI, based in Blackburg, Virginia, focuses on biocomplexity, the study of increasingly complex systems composed of many agents. Using approaches ranging from the mathematics of genetic sequencing through the dynamics of social networks and policy analysis, the Network Dynamics and Simulation Science Laboratory builds virtual cities. Based on real population data from census information, the laboratory reproduces a community, modeling a minute by minute location for each person in the city.
From there, NDSSL researchers, which include 25 staff and about 25-30 students, conduct transportation and communications studies and model how infectious diseases like influenza, pertussis, smallpox or Ebola spreads through a population. VBI has also modeled how various disaster scenarios would affect buildings, transportation, and communications, and developed a software tool enabling local health departments to use Twitter to detect food borne illness outbreaks. Similar software tools have been developed for federal agencies tracking and responding to infectious diseases.
“We have access to a tremendous amount of existing technology in traditional HPC clusters, and will use the Bright-powered big data cluster as a test bed to see how different technology might make our simulations easier to implement or speed up processing,” said Keith Bisset, a senior research scientist at VBI. “The testing will also guide requirements for purchasing a much bigger system.” Examples of projects to be tackled with the new cluster include importing infectious disease modeling code from MPI to Spark, and developing social unrest simulations using Hadoop-based clusters that take raw Twitter data and extract and code relevant information. Another project envisaged is porting graphical analysis algorithms to various pieces of an Apache MapReduce stack, including Apache Storm and Apache Giraph.
“The ability to manage Hadoop along with a new or existing HPC cluster from a single pane of glass is a capability that will be exceptionally useful to Virginia Bioinformatics Institute,” said Dr. Matthijs van Leeuwen, founder and CEO of Bright Computing. “Bright Cluster Manager manages these highly converged infrastructures and troubleshoot throughout the entire stack, all while retaining full functionality and flexibility.”
