Compuware Corporation (CPWR)
APM Customer Webcast
December 04, 2012 12:00 pm ET
Kieran Taylor - Director, Product Marketing
Previous Statements by CPWR
» Compuware's Management Hosts Application Performance Academy -- Part 1: Performance Concepts Conference (Transcripts)
» Compuware Management Discusses Q2 2013 Results - Earnings Call Transcript
» Compuware Management Discusses Q1 2013 Results - Earnings Call Transcript
Real quickly, I'll go over some housekeeping issues. All of you are currently on mute right now. Today's session will be recorded and we'll get the link through a replay, as well as link to it back out in the next couple of days. [Operator Instructions] We will do Q&A at the end of the webcast. And the way that we'll run that, if you can submit your questions using the Q&A function on the lower right side of your screen, we'll be collecting those throughout the presentation. And then we'll, again, at the end of today's presentation, step through those and get answers to them.
Next slide please. So we've taken great pains today to make sure that this is not a vendor pitch, it's not a sales pitch. That said, I wanted to spend 60 seconds just orienting you on what it is Compuware APM is all about. And then we'll cut over to our presenter today.
Real quickly, we specialize in application performance management, the optimization of applications that you improve the experience that your end users get when interacting with those key apps.
We do it in 3 areas: We have cloud performance management solutions that are really designed for those B2C public websites, optimizing their performance, understanding where third parties may contribute or inhibit your performance.
The second area is deep transaction management. Here, we're looking at individual applications and how do you optimize them. This is a particular interest to application developers, those who test in QA, app architects, that sort of thing.
And then the third area, and this is the area we'll be talking about mostly today, is what we call data center real user monitoring. This audience is really those network operations and network managers who are responsible for a wide range of infrastructure that supports a range of applications.
Next slide. So this concept of data center real user monitoring. As the slide indicates, we have a set-up where we rely on agentless monitoring devices. They're basically appliances that would sit on a SPAN or a TAP port. And they're going to look at all the traffic that's traversing the many tiers of your data center. We have specialized support for applications, as you see on the right, so that whether we're in a SAP environment, Oracle environment or we're looking at particular databases, we are able to look at that traffic on a per-user and per-transaction level. So you'll understand in particular where you might be having bottlenecks in your application.
So again, I wanted to be quick on that. Those are the 3 areas that we deliver APM solutions. And the third area, what we call data center real user monitoring, or DCRUM, will be the focus of what Mike is talking about today.
And using that as segue, next slide. We're very lucky to have Mike today. He's going to address 3 main points here: What are the big shifts in network design and architecture? What do those shifts mean for application performance and monitoring? And why the network view is critical. As I mentioned, we have an agentless approach to monitoring your application performance. And I think by the end of today's presentation, you'll get an understanding of how you can optimize end-user experience even when in these new approaches to service delivery.
So to introduce Mike Hicks, Mike is our senior product lead focused on advanced networking for Compuware APM. And you may know Mike already, he has written a couple of books on the subject, Managing Distributed Applications and Optimizing Applications in a Cisco Environment (sic) [Optimizing Applications on Cisco Networks]. Here at Compuware, we have Mike focused on all of our advanced network monitoring and application management technologies. He was originally trained as a network engineer. He's worked extensively through both public and private sectors in Europe, North America, Asia and Australia, and that's where Mike is based. He's calling in from Australia today. And without further ado, I'm going to hand it over to Mike. Mike?
Thanks, Kieran. So I want to take you through a few steps here. So, first, I'm going to outline essentially what we're seeing in the network today, before then moving on to understanding some of those network components, and ultimately some best practice techniques for acquiring that data today. So what we're seeking in terms of the network itself today is we're seeing a change. The network has been fairly static for a long period of time. But based on what's happening in terms of how the application is being delivered and being built, we're now seeing this change. And it's a forced change and it's also good for the environment. But it has presented some problems in how we actually acquire that data or use logic from a monitoring point of view.
But on the same token, it's actually made the network and is incredibly relevant in this delivery. Now, everybody talks about the Cloud Connects and the clouds out there. To actually connect those clouds together, we still need these network components. And it occupies the largest part of that delivery time from an application going from the server to the client. So, obviously, visibility into what is happening in that dynamic path is very important across there.
So we're seeing this change, I said, where we've gone from a static environment to where we've gone to the dynamic. And that's just not the fully meshed-up environment from an MPLS delivery technique. We're also starting to see that into the data centers as well. So where we've had the virtualized server infrastructure, we've also now started to get that dynamic in that virtualized network. So we're seeing switched fabrics extend. So where we have to have this multiple pass-through environment at data center for example.
This has meant that these environments have to be totally flexible. So what we're seeing, where we used to have a set and fixed in terms of number of switch ports, for example, we're now having to be able to set this switched fabric up. We have this multiple ports available, these virtual network interfaces and so on. Again, where we've had a certain [indiscernible] configuration, so from movements around, because of the applications and the way the servers are now constructed, we're starting to see this automated approach, we're starting have to force or change the environment as it goes.