Saturday, May 31, 2008

Heading for the exit!

I have just returned to the West Coast after taking some down-time during my last trip to Europe. Before getting back into the swing of things, however, I spent last weekend at Willow Springs – a well-known Californian race track. The picture I have here is of me behind the wheel during the one session I raced solo. While, strictly speaking, this was part of a High Performance Driver Education (HPDE) program supported by the National Auto Sport Association (NASA), it wasn’t long before the pace being set by the students became pretty competitive.

A few months back a friend of mine came over for diner and, as we talked, I discovered he was a real racer – competing in events around California. After getting out to the tracks and watching him race, I signed up for a couple of weekends. Today we buy cars that cannot be driven anywhere close to the limits of their design capabilities, and since there continues to be no legal restrictions to taking these cars onto the race track, I thought that it could be fun. Can I still drive smoothly, and can I avoid embarrassment alongside much younger participants?

For readers not familiar with the format of these open track events, novices (such as me) are required to participate in classroom sessions before going trackside under the guidance of an instructor. Following each session there are mandatory “downloads” where our performance is reviewed with a little education thrown in as well. It was during one of these downloads that our class instructor walked us through one of the most important aspects of racing – negotiating the corners.

The process of cornering was broken into four components; breaking, turn-in, apex, and exit. Of these, the most important component being the exit – the car will always follow your eyes - so we had to make sure we went into the corner looking for the exit and then, when committed to the corner, be looking for the next corner! As we reviewed these components, I was struck by the similarities with what we do in IT.

Breaking, and being prepared to change direction, is as much in the minds of CIOs as racers. As we survey the IT landscape, there comes those times where we just have to hold off deploying new applications and consider whether the time is right to change direction. There may have been a recent merger or acquisition where the different IT operations cannot be easily integrated. Perhaps a key vendor has decided not to support the environment, or has come out with enhancements that you cannot support. There’s always something just around the corner that necessitate a pause, whether it’s car racing or IT.

Turn-in, and knowing just when to make the decision to change, separates good IT management from those with less experience, as for many of us, determining when to change calls on skills accumulated over a lifetime in IT. You can turn-in too early only to be forced to ease up and loose ground, or you can turn in too late only to scramble to keep everything together and risk catastrophe.

Apex, that mythical point lying somewhere in the corner that immediately tells you that it’s OK to resume your attack on the next corner – that next milestone that will mandate another change. It was the great Formula One (F1) racing driver, Jackie Stewart, who on one episode of the BBC’s Top Gear redefined the meaning of the apex by instructing his student to “never press the gas pedal until you know you never have to take it off!” In other words, there comes a time when you reach a point where it’s just good to go – no more looking around and no more hesitations. And isn’t this the same in IT as we reach a point where our testing is over, where running in parallel with the old is no longer highlighting anything significant, and where we can commit to the course we decided to pursue.

Exit, the most important element of all. Jackie Stewart, in reviewing the early driving efforts of his student, observed “maybe you’re trying to think too much, about how you’re doing, rather than what’s coming up! The exit is far more important than the entry of the corner with regards to smoothness.” Speaking as a complete novice racer, this was a hard lesson to learn – early laps of the circuit had me watching the track just in front of the car and I was constantly making adjustments to the steering as well as coming off the gas peddle – I wasn’t comfortable and I certainly wasn’t smooth. Don’t we often do this, as our project enters the final stages. Don’t we become concerned that perhaps we aren’t quite ready – yet we really are at the point where we can go “live!” Shouldn’t we now be looking “through the exit” with our eyes focused squarely on the next project?

The car I took to the race track was the C6 Corvette. As the weekend unfolded I really gained a new appreciation for the car – I had many friends, particularly those from Europe, who really didn’t think that a Corvette was a serious car for weekend track days. Sure, it was fast in a straight line – but would it go round corners? They may be powerful but they are just too heavy and difficult to steer and there are many more-nimble alternatives! The cynical ones even suggested that American-built cars lacked the all-round capabilities of their more capable brethren from Europe and Asia. Could you ever be smooth in something as big as a Corvette? Could you maintain the pace?

So it was very encouraging to hear our instructor take time out to talk about safety and the need to pay attention to our seats, our seatbelts, and what we need to do to protect ourselves. “Cars built today are just so hindered by the restrictions placed on them by the lawyers and it takes a lot of effort to get them ready for track days,” he said. He then added “unless, of course you have come here with a BMW, a Porsche, or a Corvette, as these cars were developed for the race track before being refined for every day road use.” Fancy that – and who knew! It turns out that, among the racers, Corvettes hold considerable prestige and are viewed as highly desirable track day cars.

I have written previously of how difficult it is to shake off a label no matter how undeserved. And I have also observed how difficult it is to maintain relevancy when your key attributes have become “legendary.” For many, the “Stingray” suffers much the same fete as Tandem – they have a wonderful heritage but perhaps they no longer provide the right balance between power and weight, between price and performance. As the original Stingray name faded, and the Corvette brand became recognized worldwide, so too has the Tandem name, retired now following the acquisitions, become a footnote in history, replaced by NonStop. But with the resurgence in popularity the Corvette brand has enjoyed, will we see something similar happen for NonStop? As the Corvette shed weight and improved its performance, are we likely to see NonStop shed costs as its performance is improved?

In the coming days, HP will be rolling out a new NonStop product line built around commodity blades. This will have enormous impact on many market segments where previously, the thought of deploying NonStop was considered a backward step – something difficult to openly talk about with colleagues without appearing as though you were diverging from “mainstream” computing. Nothing could be further from the truth. Industry-standard chip sets, vendor commodity packaging, uniform operations and management tools, open languages and standard utilities ensures NonStop remains mainstream.

Late last year I posted a blog that I titled “Grading HP?” (December 29, 2007) where I suggested HP sales and marketing could do a better job. But in going through some of the comments posted on the blog, I came across one comment that said “So, who has failed? Have we application programmers failed to extend our applications into ways that realize the dream of the Tandem architecture? Has cost and the (purported) disappearance of (good) Tandem application programmers crushed the dreams?”

And it really hit me – as poorly as we may think HP marketing has been performing – have we stopped to think about the legendary capabilities of NonStop and to take advantage of the solutions that are out there, and available? Have we stopped thinking about writing the next application on NonStop?

I have included a second picture here – this time, it is of the former Vice Chairman of ITUG and now Vice President of CONNECT, Margo Holen – turning laps on Willow Springs. At ease with the Corvette, she turned in a couple of impressive laps and during one session, blew past a number of her fellow track day students, shredding her previous lap times and enjoying herself immensely.

Surely, shouldn’t we be encouraging our colleagues to put down a few laps with the NonStop? Preconceptions really should be dealt with and any old labels torn off as we gain in confidence. You certainly won’t need to look over your shoulder for anything from Europe or Asia - and I have to believe HP will only be even happier the more we shred the legacy labels! After all hasn’t HP, in electing to deploy NonStop at the heart of its server consolidation program, laid down the “lap times” to beat, and shouldn’t we now have this as the goal within our IT organizations?


Jim Michael said...

Brilliant post, Richard! The palpable excitement in your writing about your time at Willow Springs was great fun. Your comments about looking for the apex, and through the exit, in IT projects are absolutely spot on. Knowing when it's time to commit and having the faith in your decision, and your team, to look through the exit and begin accelerating to reach the next objective are so critical to success in IT. Thanks for a powerful analogy that I'm sure I'll be using!

It was also fun to see the pictures of you and Margo at Willow Springs as I have fond memories of the place. My sister's husband is a long time builder and racer of Formula Ford and Formula Vee cars and has raced, or crewed, at that track many times.

I remember one evening around Christmas time in the '70s when he was courting my sister; he was sitting in my folk's family room with a Cosworth crankshaft on his lap filing it to get it balanced! (then there's the time we were on two wheels cornering in an Lotus-powered English Ford Cortina because he missed a shift, but that's another story...)

I always enjoy reading your blog and this post really resonated for me. Thanks!

David Finnie said...

I agree with Jim - your analogies are spot-on (is that an Aussie phrase ? I never know these days) and insightful (as always !).

I think, in general, it's safe to say that if the application architects out there think that the target market is big enough to support their R&D, with enough left over to make a tidy profit, they will come running to a platform. Of course that also extends to seniour management within companies because target platform decisions are often made there. i.e. it is simple economics.

I think it's great that HP are moving forward with NonStop. Competition is good, and NonStop has the opportunity to provide some real competition in the enterprise server space. Time will tell if the market responds positively. If it does, we should automatically see a resurgence of R&D interest on the platform.

Personally, I would also like to see some advancements at the OS API level.

Anonymous said...

Well Richard, What a post! I have never seen such eloquent writting about track driving as yours.

As your instructor for that weekend I can honestly say, you two did an excellent job, and clearly listened to what was talked about.

I'm assuming you've watched the top gear episode with Jackie Stewart where your quoting from, and it is amazing to watch him on and off the track, a true living legend in the sports car field.

It was a pleasure having you at the track, and we should see you soon at our next event at ButtonWillow. Although I will say, your C6 will be much more challenged this time :)

Take care,


Richard Buckle said...

... and John, we learnt a lot from the download sessions so don't let folks suggest there's no value. And yes, we're looking forward to Button Willow!

.. we have watched the Jackie Stewart "instructional clip" as he helped out the Top Gear journalist (James) and suggest that anyone who has missed viewing this wonderful historical piece pull it up from you tube. Just google:
youtube jackie stewart top gear
and it's the first item returned!

All the bestm and thanks for taking the time to look at the blog ...

Anonymous said...

I do like the idea of test-driving NonStop, however I think the test-driver should be application vendors rather than CIOs or programmers. The key question will be to get the Microsofts, SAPs etc. of the world motivated to port their SW onto the NonStop blade architecture. CIOs buy "solutions" rather than "hardware".

Kind Regards,

Thomas Burg, CISSP
comForte GmbH

Richard Buckle said...

I absolutely agree with you here Thomas - the battle will be won through the application vendor wars. If they are attracted to the platform - then that will prove significant and with each win, others will begin to consider. I have been looking at this vis-a-vie SOA adoption and plan on putting something together on this in a future posting on partners, and the need for partner investments.