Improving Data Mobility & Management for International Climate Science

Use Internet2 SiteID

Already have an Internet2 SiteID?
Sign in here.

Internet2 SiteID

Providing Researchers High Performance Web Service Access to Big Downscaled Climate Data

Time 07/15/14 01:50PM-02:10PM

Room GC402

Session Abstract

As High Performance Computing (HPC) and data storage infrastructure
improves, the spatio-temporal resolution used in downscaled climate modeling
has likewise increased. This has resulted in dramatically increased
storage and network throughput requirements to deliver these large data sets
into the hands and tools of researchers. Additionally, the nature of
climate science requires significant collaboration among a diverse set
of domain researchers across academic institutions and agencies, often
within a specific geographic region.

The Northwest Knowledge Network (NKN), a scientific data management
center at the University of Idaho currently hosts and delivers dozens of
terabytes of regional and national gridded climate scenario data.
Most notably, NKN provides the storage and delivery infrastructure
for the USGS Northwest Climate Science Center (USGS NW CSC) and the
USDA Regional Approaches to Climate Change (REACCH) projects. NKN
uses dedicated file transfer equipment within our new Science DMZ
to replicate our large catalog of climate data to redundant
infrastructure housed at the Idaho National Laboratory (INL)
for the purposes of backup, fail-over, and load balanced access to
regional partner institutions.

Downloading large data sets is often cumbersome and unnecessary for
researchers. NKN exposes all of our large gridded climate data products
as published web services such as OPeNDAP and OGC (i.e. WCS, WFS, WMS),
allowing for data set aggregation and efficient subsetting. We have
also established our own best practices for handling science
granule-level and collection-level science metadata using the
ISO 19115-2 metadata standard. We will present our established
practices in handling large-scale climate data and lessons learned.


Speaker Erich Seamon University of Idaho

Presentation Media