The Future of Database Tuning and Database Administration
Let’s deviate from our technical topics for a blog or two. As database administrators, it is important for us to know what educational topics we should be spending our most precious resource on. That resource is our time. When it comes to tuning education, what should we be learning? Where should we be spending our time? Is it how Oracle shifts bits and bytes around in hash join processing or is it attempting to have a thorough understanding of the tools and advisors that are becoming increasingly prevalent in each new release of the Oracle product’?
We should all know by now that the preferred administration and tuning tool of choice is either GRID or its standalone counterpart, the Database Console. The advanced alerting, monitoring and administration features make these tools invaluable to Oracle DBAs. The advanced performance monitoring and analysis features should also make them your number one tool for performance problem determination.
The future of Oracle database tuning will be administrators interpreting and implementing the recommendations generated by the intelligent advisors and ADDM. It is a foregone conclusion that Oracle will continue to improve upon their performance monitoring and analysis toolsets. Self tuning features are no longer options that are “nice to have”, they are requirements for Oracle’s competitive survival. Microsoft SQL Server is continuing to scale, moving into areas that were once dominated by UNIX big-iron machines running Oracle databases. Oracle must compete with SQL Server’s ease of use or it will undoubtedly lose market share.
As the intelligence of the advisors and ADDM increases, the need to possess an in-depth knowledge of Oracle and the requirement to review reams of detailed diagnostics to improve database performance will decrease. And you heard it here first folks, I also think that reading SQL traces and statistics dumps will be a thing of the past. My crystal ball tells me that its just a matter of time until Oracle’s SQL advisors make SQL traces and statistics dumps less and less important until they become totally unnecessary. Tuning should NOT be a high ROI activity, it should be delegated to a mere maintenance activity. There is just too many other things for us to do. IT shops will continue to demand that DBAs be more than just technicians. Our role will be how to utilize the ever-increasing number of features provided by the database to solve business problems. NOT looking at a 47 step access path.
The new breed of top tuners will be the administrators who focus on how to use the toolsets and interpret their output. Not the tuners who spend the majority of time digging down into the dark, inner workings of the Oracle software. I’m not saying that knowing how the database works is immaterial. I am stating that this intimate knowledge will become less and less important as the tools mature. It’s only a matter of time until tuning is relegated to a minor sub-task that can be scheduled between other activities.
There will always be a breed of “super tuners” that write articles, books and speak at various events. But this segment of our population will be forced to focus less and less on the core internals of the database. The scope of the information they disseminate will need to broaden to cover new tuning toolsets and performance enhancements in the “latest and greatest” release of the Oracle product set.
If we look at a history of Oracle database enhancements, we can easily see that the vendor is identifying facets of the database that are complex or problematic and providing solutions for them. Remember manual rollback segments? How big were they supposed to be? How many? Oracle recognized that DBAs didn’t do that great of a job administering them and decided that the database was better off doing that job on its own. How about freelists? Dictionary managed extents? Multiple settings for sizing individual SGA components? Log checkpoint tuning? There is no doubt that managing the core Oracle internals is becoming easier.
I also think that Oracle will eventually become self-tuning. Personally, it can’t come soon enough for me. Don’t get me wrong, I enjoy tuning. I’m actually a fairly accomplished tuner.
But as the director of a remote services provider that manages thousands of Oracle, SQL Server, DB2 and MYSQL databases, the less time our product support teams spend on tuning, the better. If we do tune, our focus is on proactive and not reactive tuning. The remote database administration field is a tough job. Our customers hold us to an extremely high standard. I’m OK with that. They have turned over the keys to their most valued corporate assets – their data. As database administrators, we understand the gravity of that decision and we don’t take the responsibility lightly. The less time we spend on tuning, the more time we are able to spend on being proactive, using technology to solve business problems, adding strategic value, improving communication and, in general, just building a positive rapport with our customers.
And just because we won’t be spending time performing proactive or reactive tuning, we all won’t be out of jobs. We’ll just be doing different things. I have listened to various industry pundits proclaim that the next release of so-and-so database was going to no longer require DBAs for support. Bull. I knew it was bull then, it is bull now and it will be bull in the future. Database companies know that they must add new features to be competitive. Every new release contains so many new features that I feel like I have to learn to support the database all over again (which is why I like this job, by the way).
Labels:
DATABASE
