Wednesday, October 15, 2008

Who you gonna trust?

For more than a year now, I have been travelling between Boulder, Colorado and Simi Valley, California. Symptomatic of the changes taking place across our industry, where companies have become more interested in the skills and knowledge you can provide, than where you live. However, some executive positions still require a hands-on approach, and this is what has happened in our family. Frequent travel between the two cities has become an integral part of our life these days.

There’s always a downside of course – I was looking for tools the other day that I thought were in the garage, until I remembered that they were in the other garage. And when I went to the refrigerator, I remembered too late that what I wanted was in the other refrigerator. And then there are the usual household expenses that need to be attended to, along with the frustration that follows when you realize you are paying for services you can’t actually use!

The climate in Colorado means that you need to turn to contractors for tasks you would normally take care of yourself. As winter approaches, all outdoor plumbing needs to be “blown”, including all irrigation systems turned on for the summer, as water cannot be left to accumulate when the temperatures head south. Coordination between the many vendors now contracted for lawn-mowing, gardening, the plumbers, etc. becomes a highly critical task we found, as is establishing your guidelines.

And this year, I had a wake-up call about relying on one contractor looking after it all. After much higher water bills last year, and with no real idea as to the cause, I had our gardener program the sprinkler system this year. And he understood the “budget” I had for water knowing full well I wouldn’t be at home to enjoy the gardens and lawns for much of the year but just as aware I didn’t want to see them to die from lack of water. In the late spring, I began to notice the lawns were a little browner than in previous years – and that was OK. I understood. As long as nothing died!

The picture above is of the front yard with everything looking really green and very plush. At about the same time the lawn-mowing service bills began to really climb. And then I saw two months of water bills – whoa! About as much as we had paid in rent, only a few years ago! A quick phone call to the lawn-mowing service and yes, he had changed the irrigation programs – now water was coming on every evening and for a longer period. Didn’t I like how green the lawns looked?

“Never let the lawn-mowing service look after your irrigation!” was the response from our gardening contractor. “Of course they will increase the amount of water irrigating your lawns! There’ll just be more lawn to mow and they will need to mow it more often. Trust me!” So we are having the amount of irrigation dialed back as we head into the fall, and no longer relying on the services of just one vendor.

And it reminded me – how much attention are we paying to the industry consolidation going on, and how attentive are we to the mix of infrastructure and middleware software they now control? In particular, the most recent moves by the very large software vendors to flesh out their offerings so as to provide the user community with more or less a one-stop shopping experience?

Momentum first swung away from the hardware -, and operating system -, centric models of the late 60s, and has never swung back. It is very much an application-centric model today, and we acknowledge that business issues continue to determine the applications we deploy. It is the application that dictates the middleware and infrastructure requirements, and ultimately, mandates the hardware and operating systems.

Should an application only support Oracle, then there’s no point in investing in DB2 or MySQL as these middleware offerings will hold little interest for any application vendor tied to Oracle’s products! And now we watch as Oracle integrates applications, like Siebel, with former BEA infrastructure, as well as to Oracle’s data base – as it owns all the pieces and can provide a complete “mega-stack”! Likewise, as Microsoft continues to incorporate more of the DATAllegro technology (via an acquisition) into future releases of SQL Server, do we seriously expect Microsoft applications to support a raft of system platforms other that Windows / Intel? From Microsoft applications, to .NET infrastructure, to data base middleware – they too have their own software mega-stack.

Is this necessarily a good thing for a user community? Such tight integration is quickly reducing the number of vendors we can turn to for solutions and that always has to be worrying in terms of maintaining innovation, particularly where data base products are involved. Could we be handing everything over to the lawn-mowing service, trusting them not to overdo watering the lawns?

This past year, we have seen the sizes of data bases explode as the applications are capturing more data than ever before, as operational systems push deeper into corporations, that in turn, provides companies with the ability to reach out to more markets and capture even more customers. In a recent article in InformationWeek, Scaling the Data Warehouse (October 13, 2008), writer Richard Winter talks about one client where “the data warehouse keeps more than a petabyte of disks spinning and has grown by a factor of 10 during the last four years. It's expected to at least double in the coming year.” The importance of data base vendors “capturing” these applications providers, as they are now doing, becomes a lot clearer as well as a little worrisome.

And what does this mean for NonStop? I have been a very strong supporter of NonStop SQL (NS SQL) but market-penetration has been very slow to develop and now, twenty years on, I am concerned about its future. Without HP buying a very big vendor with broad market appeal, like SAP for instance, and integrating key NonStop middleware and infrastructure components such as NS SQL into these applications, maybe time is running out for NonStop!

On the other hand, many view Neoview as potentially a really big application, and its use of the NonStop data base as well as the NonStop hardware and operating systems, is a step in the right direction. In the Telco marketplace, there are significant applications from HP’s own software organization that remain very closely tied to NonStop. And some of the news I am hearing about Logica, following its acquisition of the Swedish company, WM-Card, pushing hard and fast with its NonStop “Global Payment System”, is encouraging as well. And this is what it will take for NonStop to remain relevant – many more large application vendors tied intimately to NonStop!

But should we be relying on the major vendors like HP, IBM, Sun to solve our business issues? Are we running the risk of compromising parts of the solution by ignoring better implementations for the sake of having just one vendor as our contractor? Again, are we prepared to pay the higher bills for software that will follow any deeper industry consolidation?

In Richard Winter’s article in InformationWeek, referenced above, he asks a number of very good questions:

“Do we have the right architecture? Is it on the right platform? Is the warehouse about to run out of headroom? What will it take to service new users? How do we move from batch loading to continuous update? And with technology changing so rapidly, how do we know we're on the right system?”

And then Richard responds with:

“All the answers loop back to managing scalability. Getting control of scalability might mean embracing the highly parallel processing and scale-out architectures long offered by Teradata and IBM, and elements of which are now emerging in new products from Oracle and Microsoft.”

Without doubt, it’s puzzling to read statements like this and not to read anything about NonStop and for me, the short answer to all of the questions above continues to be NonStop! But the lack of support for a HP mega-stack remains a worry for some in the user community. Or does it?

Perhaps the recent decision by HP to partner with Oracle with the just-announced Exadata Storage Server, represents the extent to which HP is prepared to go at this time. And perhaps HP is aware of the pitfalls from pursuing a mega-stack, extending from the application all the way to the metal, and understands that for many in the user community this could position HP as the lawn-mowing service that also regulates the irrigation system!

Furthermore, could heading down this path be leading back to application silos, with no options to select from best-of-breed solutions and no easy integration with competing vendor’s applications? HP, with its messages around industry standard and open solutions, surely has to be sensitive to the downside of this, and must be approaching the emergence of these essentially proprietary mega-stacks with some concern.

Yet application silos are definitely beginning to re-appear. And perhaps the die has been cast, and this will be the technology landscape for the next decade! In the work I am doing at GoldenGate, I see a huge business developing for my company in the real time as companies begin to rely more on the data as the point of integration between deployed mega-stacks.

But just as I was shocked to see the price I was paying for water – and found no easy way out of my predicament, so too, will many in the user community recognize too late the dangers from reliance upon mega-stack silos. Much as we enjoy the separation of power generation from power distribution, and the lower prices that followed, we need to be careful about not having access to the wisdom of a gardener when dealing with a lawn-mover service. It may be a grave mistake! Who do we really want to turn to? And who do we really trust?

No comments: