Editorials

Has the Tool Selection Process Changed?

This sort of reminds me of the “I don’t use the GUI, I’m a command line person, or “I use ISQL still, it’s faster and cleaner” or… yeah.

I’d have to say that, early on, I did tend to avoid overly complex GUIs and advanced tools where the information about our systems was shown first in a graphical, very pretty, very summarized way, and second in a manner that I could actually know what to do something about.

So many times, those graphical displays became the emphasis of the product. All of the development effort went into making those work better. Green lights, yellow lights, red lights – all of that.

Anyone that knows me knows I’m a HUGE fan of exception-based management of people and processes. That means I love those “show me when something is wrong” systems, but early on, they were just so summarized and so sterilized that, to me, they hindered real management of the servers more than they helped in some cases.

Yes, some great tools out there, no doubt. But I would shy away from those that were so focused on pretty interfaces and not on providing real information underlying the issues detected.

Things have changed. SQL Server and other database platforms have matured at that core level. That level that it used to be a requirement to watch over in detail. Now it’s as likely (or more likely?) that you’ll be tweaking and tuning very specific issues than overall, big, sweeping ones.

I have to admit that, now, I look for the tools that can very quickly tell me what’s up across systems, provide feedback, provide a way to quickly drill down into what’s being seen and provide actionable interfaces that make it possible to start the work of addressing what’s happening.

Before – give me a command line and an alert – I won’t have to spend time worrying about the UI to get things back in line (I’m aging myself, but you know what I am talking about, I suspect).

Now – I look for tools that provide leverage. Exception-based alerts, solutions, interaction between systems, knowing what “normal” looks like an when things hit or miss that expectation.

I won’t say I’m a full-on “just show me the pretty pictures” type of manager for systems, but I would like to have some helpful monitoring and tools that make it possible to respond quickly, succinctly and accurately, and even more, across systems as it’s likely that, whatever issue just popped up, has implications for other systems.

How do you pick tools? Has your criteria changed?