Oracle Utilities MDM is well suited for users that have plenty of money to invest in technical hardware resources to support the requirements of this application. With the electric utility industry becoming more and more precise data-wise, the requirements for data storage and processing are only going to increase. For example, electrical usage measurement used to be a once-a-month practice; in the 1990s, hourly reading was introduced; recently, 15-minute interval readings have been introduced and will become the new norm. Going from one reading a month, to 720, to now 2,880 requires expensive hardware. MDM is able to handle this load, in our experience, only if you purchase Oracle's Exadata hardware, which is priced at a premium. Beyond data storage itself, we have also found that MDM real-time usage for users is also, unfortunately, best with Exadata. Likely because Oracle developed this hardware, and because it has built-in compression, portioning, and tuning features, performance is better. One note is that Oracle does provide a purge-and-archive strategy in the more recent versions of MDM (i.e. ILM). However, implementation of this is a small project in itself, although worthwhile in the long run.
In some ways, MDM is too customizable. There is a delicate balance to be struck between allowing implementers the ability to create custom algorithms to meet their business needs and making those features part of the base product. Lots of things can go wrong if you customize it too much.
Although I have minimal experience with meter data management products outside of MDM, it does perform better than an in-house product developed by our utility 20 years ago for use on our mainframe-based system then. Meter management is headed the same way as other industries in terms of data usage and mining. The requirement for data is only going to continue to increase, and at this time it appears that MDM is only just keeping up with those requirements. Oracle has done a great job of purchasing products and integrating them into their overall framework, but the current database structure of MDM lends itself to poor performance if Oracle's own hardware is not purchased, and if their purge and archive strategy (i.e. ILM) is not employed.
Preventing errors for us always turns into money saved. Error flags that MDM throws help to prevent incorrect data entry, which almost always turns into money and time spent on sending somebody out into the field to ensure accurate data was given. So by not allowing for the possible errors, we save money.
