A federal government agency needed to move a growing 140 TB Oracle database table of image data into a system which supports horizontal scaling and can be replicated between multiple data centers. The system was required to be able to retrieve images on an average of 50 milliseconds per image under heavy load. It also had to be able to support annual growth of 10%. In addition, the existing systems that retrieve the images had to remain unchanged.
The agency selected QAT Global to provide the path to a successful system modernization.
QAT Global’s Approach
QAT Global designed a solution that used Cassandra for storing the metadata related to the associated documents and used IBM GPFS to store the image data. RESTful Services were created to interface with the different platforms for loading and retrieval of the image documents and metadata. Existing RMI services were modified internally to use the new REST services for loading and retrieval so the existing systems that retrieve the images did not have to change.