An Argument Against Big Data in IT Operations

The Big Data trend is casting a large shadow over the Application Performance Monitoring (APM) market right now. Most vendors have great solutions for generating Big Data (that’s the easy part, right?). A few vendors have solutions for analyzing Big Data. But, as far as I know, we’re the only one generating AND analyzing Small Data.

That’s right, I said Small Data.

Small Data consists of simple, contextual facts that help you complete a task faster or better. Big Data consists of intricate, thought-provoking insights that help you discover new opportunities and innovate. Everybody needs Big Data sometimes, but most of the time, most people need Small Data.

Take IT operations teams for instance. Every once in a while you need to plan capacity, explore better ways to run applications or infrastructure, etc. That’s a Big Data problem. Keeping it running smoothly and efficiently day in and day out? That’s a Small Data problem.

However, most IT operations teams I talk to have a Big Challenge. On a daily basis, they have to turn Big Data into Small Data to do their jobs. It’s hard to do and takes a lot of time.

Because they have so many monitoring tools, they end up with Big Data. Terabytes of application and infrastructure performance data chock-full of interesting insights. But, they’re trying to complete very specific tasks like “figure out why this service is running slowly, and how much of an impact it’s having on the business so I can get the right team working on restoring it to normal service levels.” So, they start downsizing their data to bend it to the task they need to complete. Big becomes Small through significant human effort.

INETCO Insight is built around a Small Data philosophy. Simple, contextual facts to help IT operations teams complete tasks faster. You’ll only find five metrics on our dashboards: duration, rate, failed rate, network delay and application delay.

application dashboard1Within a few clicks you can get these measurements for any application, transaction, component, or network link. You can see them in real-time, over time, or go back in time.

online banking rates1When you see a data point or trend you’re interested in, you can get a short list of the exact individual transactions generating those data points. Each of those individual transaction records is presented exactly the same way, with instant access to complete details on the transaction (This is one place where we have a LOT of data, but because it’s all about a single transaction, it’s not Big Data).

e commerce transaction1Don’t get me wrong, INETCO Insight isn’t a simple piece of software. We actually solve a whole host of ugly Big Data problems so we can generate and analyze the Small Data IT operations teams need. I think we’ve got it right – Big Data is for computers, Small Data is for people.

If you are interested in learning more about the INETCO Insight application performance management software, you can download this product overview whitepaper, or contact INETCO for a demonstration.