Last month I wrote a guest musing and I have to admit that I rather enjoyed it.  I received good reaction to it and got a chance to connect with some interesting people as a result.  It seems as though the closet-cleaning ideas “struck a chord,” as one person told me.  Well, I don’t know if Mark is just too busy to stop me or whether he actually enjoyed reading about network management issues (would he admit it even if he did?) – but when I mentioned that I wanted to write a follow up he agreed.  This month’s musing begins where last month’s ended.  In other words, I want to discuss where network and systems management are heading.

As I write this the stock market indices are down to five year lows.  All you hear is the doom and gloom in the market.  I have nothing of value to add to a stock discussion but I can tell you how all this news makes me feel.  You see, I am an incurable optimist who expects things to turn around.  So now that we have had a chance to deal with what has happened in the market and with the economy, I am left with a strong desire to clean house in preparation of better things to come. 

A couple of month’s ago Mark wrote about the concept of Thrivers and Survivors in these difficult times.  I believe that management companies have great potential to be among these Thrivers and Survivors – but they have to stick with some very basic value propositions to the IT Operations Manager:

1. Reduce costs both (capex and opex).

2. Make what I already have work (like I was promised at the point of sale).

3. Help manage some of my new problems.

4. Provide a product specifically useful to me and my organization, not partially to me and partially to others who I don't share budget with and can't plan with.

5. If there is any confusion . . . see rule #1.

I think the time is right to re-evaluate some of the ways we have managed networks and systems.  This re-evaluation should be done with a goal to keep what really provides value and shed what does not.  No idea or issue should be immune from consideration. What should emerge from this are the areas of opportunity for emerging companies with new technologies.  Not only will these companies be poised to fill clear “gaps” but they will be coming of age when the real buying starts to begin again in earnest.

So what’s important when we manage networks and systems?  Well, first we have to configure components; next, make sure they are running and then monitor to see when and if we need more.  In these times it’s also critical to make sure we protect our resources from unauthorized access.  So, it’s only natural that the priority order in management has been Configuration, Fault, and Performance and now there is a strong push for Security.  Of course, if we’re in the service provider market, accounting must be covered.  But in the enterprise, it’s not that important so far.

To date, most networks have done a pretty good job of covering these basics.  Because of the critical need for fault management, the different systems – performance, security, and element managers – are interoperating with the fault managers.  With the rise in prominence of fault managers, we may be seeing the beginning of the end for frameworks and platforms.

Over the last six years one of the difficult things I have had to do with clients is to try to explain the purpose of the framework.  The framework vendors have had an opportunity over the years to exploit their strong position in the NOC infrastructure, but they have largely fumbled the ball.  Going forward, they will have to do a better job of making disparate systems work as a system or face becoming irrelevant.  Is it too late for frameworks?  Probably not.  But they will have to figure out how to add value without the complexity that is assumed to be required in order to achieve their value. And it will have to be done at a time when most of the compelling issues involving node-based infrastructure management have been done.  There might be some minor problems to be solved but nothing big – mostly they’ll have to figure out how to integrate systems better or fade out of relevance.

But there are some emerging areas of management that are gathering some interest – the activity is moving away from the network nodes and toward a number of overlooked areas that make applications work.

Here are a few areas that are rapidly becoming critical to IT managers.

Storage and Data Management – We’re on the verge of a revolution with storage – moving from server-attached disks to storage networks that store and archive data as a utility for the applications and the corporation.  This revolution will encompass the servers, spindles, storage infrastructure, and the management of those systems. Add to this the increased focus on the criticality and security of the data that is maintained in these storage systems and the stage is set for new systems to simplify this potentially complex and very important area of the enterprise system. This need is real with top tier enterprises and will only grow throughout the market.

Business–Aware Systems Management – A variety of start-up companies are addressing this growing need – specifically the desire to pull together all of the information in the enterprise and reflect health not in terms of network infrastructure but instead in terms of what matters – the business processes.  These processes are what motivated enterprises to spend millions on applications and networks.  IT managers are now rightly asking to be able to manage at the business process level – not at the node level.  This makes sense to me – and these start-ups are poised to take advantage of this need.

Application Management – The place where you draw the “cloud” in network drawings has moved from the LAN/WAN infrastructure to the application architecture.  Business-Aware Systems Management tools will help us identify where the components of the problem are but there is a need to dig deeper into the transactions. Over the past couple of years corporate enterprises have implemented response time monitors as a way to better understand the end user’s experience with the application – only to find out that when the server was the problem, they didn’t know where to look.  The “server” is a multi-tier, multi-dimensional “cloud” that needs to be viewed and managed as a mesh of logically interconnected nodes and transactions.  There is a growing need to understand the behavior of these components at the transaction level – monitoring, diagnostics and planning.  Tracking transactions from the client through this application cloud is a complex problem and companies are only beginning to try to tackle it.  The good news is that when they do they will find a large and appreciative audience among Enterprise IT managers.

Service Provisioning – Currently service provisioning – linking simplified service creation, end user control, service level monitoring, and any billing or chargeback – is restricted to the service provider.  Bringing service provider-like provisioning to enterprise applications, plus managing and securing the application “cloud” will be high on the priority list for the next few years.

OK, so these seem like compelling ideas, you say.  But how can a management company make money now when they haven’t had that much success in the past?  There are a couple of reasons.

First, we’ve had a couple of decades to work at it.  We have exploded the mainframe and spread it out in the form of LANs and WANs.  It’s hard to effectively manage a moving target, but more and more enterprises leverage common technologies and best practices that give the management vendors a bigger and more stable target to shoot at.

Second, instrumentation is more consistent and better understood.  Through trial and error, we have achieved a basis of understanding for the amount and types of data that should be gathered to achieve specific goals.  We also can more surely depend on the instrumentation being available in the majority of devices and systems – thus assuring the widest possible success of an idea that is dependent on instrumentation sources.

Third, and possibly most important, a growing understanding has developed so that the modern-day corporation understands that their applications are at the center of their corporate effectiveness and possibly even survival.  This has created urgency for better management of all components that make these applications work – storage, business processes, provisioning and application management.  This urgency further strengthens the IT manager’s willingness to pay and their desire to seek out products that solve these emerging problems.

So, without question, there are network management Thrivers and Survivors now and more will emerge in the next couple of years.  But these companies must figure out now what role they will play in the new Enterprise or Service Provider NOC and not be afraid to rethink old assumptions or even throw out the old things that don’t work (or are too expensive to make work correctly).

Providing real value will be the only thing that matters – not sex appeal but real substance.  If you can’t do this, it won’t matter what your product does or how long it’s been around – it will be tossed in with the other discarded items from the closet.