MDM - If You're Willing to Pay the Price
Pros
- Integration-wise, MDM syncs with the OU product CCB without issue.
- The documentation behind the functional use of MDM is well maintained. This helps functional analysts when business issues arise and DBAs for technical issues.
Cons
- Integration-wise, MDM is flawless with CCB, but the amount of time it takes to load and interpret daily reads is nearly prohibitive. Through a process called "meter interrogation," Oracle is supposed to process initial reads into final measurements. This is supposed to run three times a day, but because the process is so resource-intensive, we are only able to run it barely twice a day. Production resources lag during this time and user experience is reduced.
- The amount of data the MDM require to be kept in just two of the thousands of database tables indicates a very poor design, or at least a poor integration of the Lodestar product that they purchased and turned into MDM. The initial measurements and final measurements tables take up around 85 to 90 percent of the database. This bloatedness in database size translates into slower performance for the front-end user as well as real costs in terms of data storage. Any user using Oracle's Exadata software will pay dearly for not having a purge and archive strategy.
Return on Investment
- Our former mainframe system would not have been able to handle the new requirement we have in Pennsylvania for 15-minute interval reads. MDM being able to handle this with base functionality saved us the cost of developing a custom solution for this.
- MDM's hardware requirements are quite expensive, especially if you want the best performance technically and for the front-end user, as you must purchase Oracle's Exadata hardware. This capital expenditure has been heavy for our utility and supported by increased tariff measures.
