End-to-end application management - the network perspective

What if your executive team needed to develop a true "end-to-end" application management strategy? What would it look like?

You might believe, and with some real justification, that the term “end to end” is only used by vendors who custom-fit the definition to the scope of their particular product.

Does “end-to-end” application management, for instance, include the mainframe? You bet it does if you’re a vendor that manages the mainframe environment! Does it include capturing the end user experience at the end station, desktop, or mobile device? Once again, the answer is a definitive “yes” if you’re a vendor that has strong QoE (Quality of Experience) roots. Or how about insights into the code and design of the application itself? If you’re one of the few vendors that does this, you’re proud of it and wouldn’t have it any other way!

But what if you were to stand the question on its head and ask: What if your executive team needed to develop a true “end-to-end” application management strategy? You would rightly assume that no one vendor is optimized to do it all. And you would be right in realizing that really effective application management requires teaming across various constituencies – application support, data center, service desk, application development and network, as well as business planner and consumer constituencies. In parallel, you would need to combine on vendor architectures and technologies “across silos,” as well.

So if you did all that, then what would it look like?Well, EMA decided to take on this challenge and collectively (working across practice areas I might add) came up with at least our first iteration of an “End-to-end Application Roadmap.” We tried to define the major technology spheres relevant to this end-to-end challenge. Then we followed it up with some research across a total of more than 400 respondents in two surveys in June of his year that paralleled some complementary network management research in the same month. All this produced what I think are some rather interesting results.

So what are the technology classes that apply to our “end-to-end application roadmap?”

They include:

• Application Integration Analytics – This area is something of a black hole still today. These technologies look at interdependencies across applications and middleware with variable transaction paths. This includes everything from capturing the dynamics between DNS and Microsoft Exchange, and Web 2.0 and SOA application components that may interact in real-time manners, to running requests through MQ Series as an extension of a total application ecosystem.

• Networked Application Management – This area focuses on application flows, including transactions, across the networked, distributed infrastructure and is one of the most active areas of innovation in the market today. It includes a wide variety of technologies to capture performance and response time, as well as volumes and routed directions, to looking potentially at transactions between the data center and the end station.

• Application Code Analytics – This sector of our roadmap targets components within specific applications. It is especially critical for what our data shows to be the number one requirement in many environments – in-house developed applications which don’t come with their own instrumentation and best practices. But it’s important for capturing application behaviors overall.

• Application Dependency Mapping – This is still a new market that helped to generate a lot of interest in, among other things, CMDB System deployments. Application dependency solutions typically provide reasonably dynamic insights. Updates may occur on a 24-hour basis in most cases, or else in "real-time" as changes made to systems-to-application inter-relationships trigger updates to the app dependency software. Most of the vendors in this market, but not all, have been acquired by platform makers. As a group they represent a wide range of technologies, and some include networked application flows with application volume awareness.

• Data Center Analytics – I didn’t name each section as this was an EMA-wide effort. And if it were up to me I would call this “infrastructure management.” It includes many of the most established capabilities for looking across the distributed infrastructure – systems, network, database, etc. – to manage the physical components for performance and availability in terms of historical trending and in real time. Since application performance depends in large part on the health of the infrastructure underneath, this is very much a part of the big picture.

EMA collectively – not just me, since many of you know I’m an evangelist – recognizes that truly effective end-to-end application management depends on a federated CMDB System underneath these various technology areas. The CMDB System can provide a cohesive set of insights into how changes impact service performance, while reflexively helping to troubleshoot difficult problems by looking back in time across relevant change histories. This, of course, is a larger discussion, but much to my delight I’m seeing more IT organizations recognize that CMDB adoptions are catalysts for both organizational and process transformation. CMDB Systems are also about architectural integration -- to optimize multiple management investments so that they work more effectively together.

In any case, we asked our respondents across all of IT to vote on all the top five categories (we didn’t ask about the CMDB as a separate option in this question). Guess which won as “most critical to end-to-end application management in 2008”?

The answer is “Networked Application Management” – that flow-based look at end-to-end transaction capabilities. It is an area with critical vendor innovation but still a source of some confusion indeed. As examples, a short (alphabetized) list of some of the more interesting choices (not including platform or major infrastructure providers) include AdventNet, Apparent Networks, Coradiant, Fluke Networks, NetQoS, NetScout, OPNET, Packet Design, PacketTrap, Valencia Systems, and Wild Packets.

In my next two columns, I’ll be looking more in depth at how this networked application market is evolving, along with sharing some of the other key “lessons learned” from our June research.

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.

Copyright © 2008 IDG Communications, Inc.