Sometimes I am asked for my opinion on the future. You know, something like “What do you see as the future of database and data management technology?” or “what will be different 5 or 10 years from now?”
I try to avoid answering these types of questions because the truth is, nobody knows the answer. If I could give 100% accurate questions about the future I would have long ago cornered a market and retired.
But I can give a general answer to this type of question. My basic answer is that, at a high level, things will be getting more automated, more connected, more wireless, and more complex.
The current wave of database administration is intelligent automation. Note that I said intelligent automation, and not simply automation. I refer to this as implementing software scrubbing bubbles that “work hard, so you don’t have to.” (Remember that commercial!) DBMS software and third party database solutions are providing built-in intelligence to begin performing tasks that, in the past, required human intervention; that is, required a DBA. As these features become more robust and functional, the DBA will be able to turn over proactive management of certain maintenance and administrative tasks to the tools – leaving the DBA more time to be proactive himself.
With intelligent automation the software becomes smart enough to figure things out without human intervention. So, instead of the DBA needing to constantly monitor database statistics to determine when to reorganize a table space, the software will keep track of the statistics and have built-in knowledge of what those statistics mean. And we will see more knowledge being applied than is currently happening. For example, one of the indicators for reorganization is cluster ratio. If the ratio falls below say 90%, then think about reorganizing. But software with intelligent automation can take advantage of performance management software that will examine access patterns. Say for that same table space 98% of the access is random (as opposed to sequential). This means that cluster ratio is not as important, so the “scrubbing bubble” can factor that into the mix and schedule the reorganization based not just on statistics, but actual usage patterns. This will be revolutionary.
Furthermore, the software will understand the mitigating circumstances, such as “Can’t schedule a REORG in the middle of the day when database activity is high”, “Have to work the REORG into the batch window,” and “These particular jobs will highly benefit from a REORG so is there room to squeeze the REORG in before that.” This will be done by tighter integration between database tools and systems management tools and by intelligent analysis of job execution trends.
And if you want to take things a step further, eventually databases will continuously reorganize themselves during periods of downtime based on statistics and trends.
Of course, some have been working on this type of software for a long time, and it will take more than 5 to 10 years to get it to the point where a DBA trusts it without constant surveillance.
Regarding the wireless aspect of things, pervasive devices (mostly smart phones, but also tablets) will increasingly work with database systems. The difficult aspect of this transition will be the process of managing the synchronization of the remote device with a centralized database server. DBAs will need to get involved there to ensure successful data synchronization. Another aspect of the wireless trend will be wireless DBA – where the DBA uses a tablet or smart phone to check on the health of the databases. This will be another tool in the DBA’s bag of management tricks – but the DBA will not be able to perform all of his tasks using a smart phone… just those that require round-the-clock vigilance such as checking to see if the database instance is available or checking the status of system parameters.
What about complexity? Well, consider the fact that most DBMS vendors update their software every 18 to 36 months. And that means more features and functionality to learn about and use. Consider also another big database trend: the DBMS encompassing technologies and functions that previously required you to purchase separate software. Remember when the DBMS had no ETL or OLAP functionality? Those days are gone. This will continue as the DBMS adds capabilities to tackle more and more IT tasks.
So, as a data professional, when you think about where you want to be 5 to 10 years from now, consider these high-level trends and what you can do to make yourself more aligned with them. To repeat, things will be getting:
- more automated,
- more connected,
- more wireless, and
- more complex.
And while you’re at it, learn more about your company’s business. The more you know about the business, the more valuable you will be… good luck!