Getting the data in (and out)

Getting the data in (and out)

Behind the scenes at GRN we have an extensive database of languages, recordings and audio files. Our main servers in USA and Australia have more than 55TB of storage capacity, and together hold over 1.2 million audio files. It's a lot of data to look after.

We have developed an in-house application to maintain all this data. It's called GRID - the Global Recordings International database.

GRN staff from anywhere in the world can access the main database, as long as they have a reasonable internet connection. However many of our centres, particularly in Africa, simply don't have good internet capability, so that's a problem for us.

GRID is written in Embarcadero's Delphi Rapid Application Development environment. Data is stored on our servers in the open source PostgreSQL database, which has no trouble with large and complex data.

To connect the two together, I've recently been experimenting with the PosgresDAC Data Access components from MicroOLAP. This it proving to be faster than our previous ODBC connections, and easier to deal with, so it will help speed up GRID particularly for our centres with poor connectivity.

GRID also stores a subset of the main database on each user's PC, so even if they don't have any internet connection, they can still do research and run some reports. This is a great asset to our workers in less developed countries, and for any recordist who travels.

It's a tough job collecting and disseminating data and audio to operations in 40 countries, but GRID is a key part of it.