Supermarket chain Morrisons is an Oracle shop, using a variety of the firm's products including Enterprise Linux, BIEE (Business Intelligence, Enterprise Edition) for reporting, Oracle Data Integrator (ODI) and real-time data integration tool GoldenGate. Speaking exclusively to Computing, Jonathan Walsh, MIS manager at Morrisons, described the five-year replatforming exercise that his firm is currently just over half-way through.
"We're in the process of a big replatforming exercise from a mainframe to an Oracle footprint. The core transactions are now going into the new data warehouse, and as the system continues to be rolled out, new transactions will go straight into it."
Walsh said that the decision to replatform was taken four years ago, as the previous mainframe environment was not fit for purpose for the size of company; it failed to scale with the business.
Although Oracle has recently announced the new version of its database appliance, Exadata X3, Morrisons is currently in the process of implementing the older iteration, X2.
"We've just gone live with X2, it gives us fantastic levels of data compression. We're seeing 20 times compression, so we go from 20TB to one. X3 however can put your entire warehouses in-memory," he added, although admitted that there are no plans as yet to upgrade to the newer model.
The reason for choosing Exadata, according to Walsh, was performance.
"Some queries used to take 10 minutes with the previous platform we were on, and that's now down to a few seconds."
But how does this vastly increased performance help the business? Walsh said that it creates a different way of working with the data.
"People used to be inclined to take the data out of the database and put it into a big spreadsheet and work with it there, rather than using the power of the platform [because queries took so long]. Now they use the system for the purpose it was designed for."
While this implementation has been a success, there is still a lot of data migration activity to complete.
"We're on a journey," said Walsh. "We've got a lot more data to put in, including years of basket-level transaction data. There's a lot of value in that data, we've just needed a platform to allow us to leverage it."
[Turn to next page]
By eliminating high entry costs for big data analysis, you can convert more raw data into valuable business insight.
A discussion of the "risk perception gap", its implications and how it can be closed