• June 2013
    M T W T F S S
    « May   Jul »
  • Advertisements

Running Stochastic Models on HTCondor

HPC Wire (05/30/13) Ian Armas Foster

Research by Brigham Young University’s Spencer Taylor applied the open source HTCondor software to a water resource model called Gridded Surface Subsurface Hydrologic Analyst (GSSHA), which requires computationally intensive stochastic functions found in many scientific fields. HTCondor executes jobs on a local network by harnessing computing power from idle systems. The resulting tests demonstrate that HTCondor is viable as an alternative to obtaining extra high-performance computing (HPC) resources for mid-level research institutions. The purpose of the project was to enable such institutions to combine their computing base with existing HPC resources both on site and in the cloud. “We found that performing stochastic simulations with GSSHA using HTCondor system significantly reduces overall computational time for simulations involving multiple model runs and improves modeling efficiency,” Taylor contends. He says HTCondor’s inherent nature required each stochastic model to run on a different number of processors, ranging from roughly 80 to 140. “As expected, with about 100 times the computational power of normal circumstances, I was able to essentially reduce the runtime by factor of 100,” Taylor reports. He also cites the possibility of utilizing commercial cloud resources as part of the HTCondor base.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: