Seamless Access to Data From Multiple Data Centers Through Federation

Date: 4/24/2019

Time: 05:00 PM

Room: Elliott Bay

The IRIS Data Management Center (DMC) is the largest public repository of passive- and active-source seismic data in the world with over ½ petabyte archived and growing. Approximately that same volume of seismic data, from different stations, is publicly available from other data centers distributed around the globe. Until recently, the discovery and collection of data across centers required researchers to contact each center individually, often using different mechanisms. In 2013, the International Federation of Digital Seismograph Networks (FDSN¹) adopted a web service specification for access to data in standard formats. In the years following, multiple data centers have implemented these interfaces. Leveraging this uniformity, we have created the IRIS Federator to integrate federated data discovery and access into our tools. The public interface to the Federator is the irisws-fedcatalog² web service, which provides the capability to search for time series data across multiple data centers. We initially integrated federated data access into our FetchData3 and MATLAB³ tools and the community maintained ObsPy⁴ Python framework. We have more recently integrated federated access into Wilber⁵, a web application for selecting and downloading data event-oriented data, the Metadata Aggregator⁶, a web application for browsing metadata and GMap⁷, a station mapping service. Ultimately, these efforts will make more data more discoverable, accessible, and usable for all researchers.

¹International Federation of Digital Seismograph Networks: http://www.fdsn.org/

²IRIS Federator, irisws-fedcatalog web service: https://service.iris.edu/irisws/fedcatalog/1/

³IRIS web service tools: https://service.iris.edu/clients/

⁴ObsPy, Python framework for seismological data: http://www.obspy.org/

⁵Wilber, event-oriented data access: https://ds.iris.edu/wilber3

⁶Metadata aggregator: https://ds.iris.edu/mda/

⁷Station mapping service: http://ds.iris.edu/gmap/

Presenting Author: Chad Trabant

Authors