Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Why every performance tool stinks

Craig Shallahamer, who makes a living planning and optimizing Oracle-based systems, thinks every DBMS performance tool sucks, including his own. Here's why.

Craig's Corner
Insights on Oracle technologies and trends by Craig Shallahamer
About the Author

Craig Shallahamer has 18-plus years experience in IT. As the president of OraPub, Inc., his objective is to empower Oracle performance managers by "doing" and teaching others to "do" whole system performance optimization (reactive and proactive) for Oracle-based systems.

In addition to course development and delivery, Craig is a consultant who was previously involved with developing a landmark performance management product, technically reviews Oracle books and articles, and keynotes at various Oracle conferences.

To view more of Craig's work, visit www.orapub.com.

You may be wondering why someone who makes a living planning and optimizing Oracle-based systems thinks every performance tool sucks. I'll tell you why! I'm surrounded by performance problems and performance tools that suck... including mine.

I like my tools. I really do. They are simple, fast, and do the job. They have evolved over the years into a very complete and advanced tool kit that I can quickly install and use. Like any other tools, they are limited and could be improved to increase my productivity and creativity, but that takes time, which is something I closely guard. So, until I find something better, I'll stick with what works and watch for the system management tool industry's new wares.

I don't think I'll see anything anytime soon. The problem is the industry. With all its millions of market research, it still has not captured the true need and the true heart and soul of a DBA. There is so much information oozing out of performance tools, someone who can read The Matrix language still could not understand this stuff. I know the information is there in some tools. But, I'm also equally certain the information presentation is not. And so, we enter the world where over one hundred performance tool vendors compete for your IT budget.

With all the tools on the market today, you would think a problem's root cause could be firmly established within four mouse clicks. But most software wants you to look at tons of information and sift through all the sludge. It's like using an email filtering program...you still have to double check all the filtered emails just in case -- worthless. How about a lot of intelligence built into the product? I'm talking about automatic, self-aware, and root cause identification. But I don't think there is anything on the market yet. Vendors will argue this point, but I'm not convinced. It still takes a skilled professional to sift through the data and a highly skilled salesperson to make you believe it's automatic and self-aware.

And don't tell me that Response Time Analysis (RTA) is the answer. I published the first RTA paper, and I'll tell you it's just a piece of the pie. A significant piece, but all the same, just a piece. It's an answer, not the answer. Some professional RTA tools want you to do absurd things to your system (which I won't mention) so you can diagnose the problem. Nope -- not going to do it. There has got to be something better.

Something like graphics that actually mean something. You've been there too. You're standing around a really slick graphic depiction of a system and after a few hours someone has the guts to say, "What's it mean?" Another hour or two later someone else says, "I don't know, but it sure is cool looking." Cool is not good enough for me. I'm looking for graphics that grab my soul and bring me into the system, while I still maintain and even flourish in my humanity.

I know our industry's companies can do better than striving for the next quarter's net-profit-deductible-after-tax-net-gain-nothing. They just stick with what "works." Does this sound familiar? They find salespeople who can still sell an 8-track tape player, give them a significant percentage, make sure they aren't familiar with words like Enron and ethics, help them find big IT budgets, and then let them go to work. It sounds crazy until you actually meet salespeople like this. Honestly, take a close look at those stacks of binders from really expensive software that is supposed to make your job easier and more fun…they never get used.

The software is not used because the solution is not elegantly simple. I mean really simple and really elegant. The first user interface for my rightsizing product, HoriZone, was very thorough, but it took people too long to figure out how to use the system. So hardly anyone did and I wondered why. Then I ran across a book that asked the question: How would Leonardo DiVinci use a computer? After reading the book it dawned on me why I had this love/hate relationship with Windows, why people weren't flocking to HoriZone, and why the best performance tools on the market today do, in fact, really suck.

It's because our tools don't do what we want them to do in an elegant fashion. Most tools try and conform us to them, not empower us. Ask yourself, when was the last time your computer really empowered you to create something beautiful something that you're really proud of? For most people, it doesn't happen very often, if ever. And that's sad. We all have the innate capability to create beauty. But the tools we use just help us get by, not fly.

I know I'm whining. But as an experienced performance specialist, and especially as a consumer, that is my right. I will always be looking for the DBA tool that will expand my capabilities and allow me to leave work feeling good, not tired.

Someday I'll see that tool... I hope. I haven't yet. But when I do, I'll post it on my web site, use it on all my performance engagements, and sell it with a passion.

Dig Deeper on Oracle database performance problems and tuning

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.