THE IRIS DATA MANAGEMENT CENTER: ENABLING ACCESS TO RELATIONAL DATABASES USING DISTRIBUTED INFORMATION TECHNOLOGIES

The IRIS Data Management Center (DMC) in Seattle, Washington is the host for the largest seismological database of its kind in the world. The DMC relies heavily upon Oracle for its relational data base infrastructure. During the past several years, emphasis has been placed upon the development of robust and effective technologies that allow access to the information at the IRIS DMC in a variety of ways. This talk will highlight the general structure of the IRIS DMC and the method we use to manage the gigabytes of information in Oracle as well as the tens of terabytes of time series data we have in our mass storage systems.

An overview will be given of the methods through which users can access the information in our relational database systems as well as the mass storage systems in order to conduct their research. A discussion of three distinctly different approaches will be presented. These include an email-based method, a web-based method as well as a multi-tier distributed computing environment we call the Data Handling Interface (DHI) will be presented.

The DHI is based upon Common Object Request Broker Architecture (CORBA) and provides direct connections between client applications, normally provided by IRIS, and data center servers established at several data centers within the United States. The DHI presently supports the distribution of information about seismic events, information about the seismic stations that record the time series and finally the time series (seismograms) themselves. The system is capable of transmitting seismic waveform data in real time. The system that currently exists provides the infrastructure needed to develop a distributed system of interconnected data centers using the DHI framework.

IRIS has plans to extend the DHI framework to other disciplines such as geodesy. We also intend to build bridges between the DHI and other large data integration activities such as the GEOsciences Network (GEON).