Operational Analytics — Business Intelligence Tools | HP

So Chris one of the hottest areas inside business service management right now is operations analytics I mean big data we hear a lot about but it’s not just for CFOs and CMOs is it. No if you think about IT today it’s really a problem around massive performance data, event data, log data and trying to turn that into actionable insight to help IT run their processes more effectively. Now you and the team have just about to launch a new solution, can we take a look? Yeah let’s take a look; what we’re looking at here is a solution where we bring together log analytics, machine data and our event and performance data. Were in our operational cockpit our application monitoring has identified an application is running slow which is correlated to a CPU bottleneck on a particular server. Wow that’s great and of course that correlation is one of the reasons why business server management from HP is one of the market leaders. Now typically what would happen once you identified that sort of a problem? So what would normally happen at that stage is maybe an application owner or a support person would start logging into each of the systems that were involved in that application and start to look at the log messages. Sounds expensive. It is and it’s and a common challenge that our customers have. What we’ve done here is in the context of that particular server and that application we’ve stored the log files from all of the application servers that were involved in the application as well as the event history all within the logger. Let’s take a look. So we see lots of events that are on the CRM system that’s the system that’s having the performance issue but are traffic over time doesn’t look to be too bad so it’s probably not a peak in traffic so let’s look at the machine data the log messages that are being created for that application. I’m not seeing any error messages, I got a few warning messages from the log file so that doesn’t seem to be a problem but I’m seeing a large number of debug messages on this graph. Which you wouldn’t expect to see on a production app right? Certainly not so that’s probably some issue so will drill down one level deeper we can see the amount of times those log messages have been written through the histogram and we can see were getting multiple debug messages you know multiple per second for this particular application. Looks like you found our smoking gun Chris what happens next? So typically what you would do is back within the operational cockpit within the operations bridge you would then follow your normal processes, assign it to a user, create a ticket within your incident management system or even collaborate with the application support person or developer and get them to change the application right now to get the performance back to normal. Using enterprise collaboration. Exactly. Fantastic so we’ve been able to consolidate multiple point tools we’ve been able to save obviously a lot of time and expensive effort and we’ve been able to do all of that without changing anyone’s workflow. That’s correct. Fantastic

About The Author
- HP creates new possibilities for technology to have a meaningful impact on people, businesses, governments and society. With the broadest technology portfolio spanning printing, personal systems, software, services and IT infrastructure, HP delivers solutions for customers’ most complex challenges in every region of the world. More information about HP (NYSE: HPQ) is available at http://www.hp.com

Tell us what you think...