Skip to content

Atlantis, Discovery and Endeavour…The “Last Rides into Space”

May 27, 2010

Early on in my career as a high tech salesperson working for IBM in Los Angeles in the mid-1970s, I had the good fortune to do a walk-through of a mock-up of the Space Shuttle.  My first impression was:  “Boy, is this thing SMALL!”  (I’m confident that NASA has enough sense to test astronaut candidates for claustrophobic tendencies.)  Believe it or not, at this time the nascent aerospace business accounted for nearly 25% of all employment in the Greater LA area.  Since then, most of the big players have re-located to other geographies other than Southern California.

As a young computer salesperson, especially from “Big Blue,” I was eager to ask many questions about the information systems being proposed to propel this “flying brick” into space AND return it to earth…in one piece.   I was told the plan was to have FIVE (5) separate mini-computers (mainframes were a BIT too large…) do the wildly complex computations required to pull this feat off successfully.  (BTW: the GRiD Compass laptop, at $8,000 in 2008 dollars, was also onboard the initial shuttle flights as well).  Four (4) of the 32-bit minis were to be redundant, in “lock-step” with each other, with voting algorithms used to identify and isolate any single computer that failed, either hardware or software.  The original computers had no hard drive and loaded software from mag tape cartridges.  Don’t laugh, even after upgrading the hardware in 1990 (wow – twenty years ago), each computer has a grand total of about one (1) Meg of main memory capable of a mere 1.2 MIPS each!  Oh yeah, they also upgraded the memory from magnetic core (!) to semiconductor with battery backup.  The computerized keyless fob that operates your late model car probably has more compute power, and for sure, today’s cars have more computer power than the Lunar Lander. The fifth computer was there for backup purposes only (NEVER engaged, in 132 missions so far…thank goodness!), with a completely different operating system and completely different application and communication software as well.  This redundant computer architecture (both hardware and software) preceded what we all now take for granted in Computer Land:  LANs, multiple high speed data busses, “fail safe” software (apps + OS + comm, etc.) and a myriad of other modern day high tech standards that keep our companies and economies running (for the most part) smoothly.

This “back up” plan made me extremely nervous (I have always wondered if ANY of the NASA astronauts who actually flew into space were computer scientists…”Ground control to Major Tom…”) as there is zero statistical probability that the backup computer would be able to replicate the four redundant “lock-step” computers 100% of the time if for some reason, all four of the operational computers failed at the same instant. With over 2.5 million distinct physical parts, each of the six Space Shuttles are considered to be the most complex “systems” ever produced…so far, and as the old joke went, with each part supplied by the lowest bidder.  BTW, these numbers do not even include all the software that is required to drive the 4.4 million pounds of technology at 17,320 mph into space and glide back through the atmosphere at 17,000 miles per hour down to a ‘mere’ 330 mph before final approach.

My point here is that even with some of the brightest engineering minds on the planet (and we’re not including our Russian friends…) figuring out how to pull off 132 missions with only two mishaps (Challenger/1986 and Columbia/2003) is nothing short of a miracle.  Recent estimates put the total cost of these “miracles” at over $170 billion in 2008 dollars.  In fact, each launch alone costs out at nearly $450 million per attempt or roughly $1.5 billion per mission.

Without the underlying IT infrastructure, none of these “miracles” would have happened.  Remember, these space vehicles were the very first to deploy fully computerized fly-by-wire flight control systems.  This means that the shuttle pilot’s control stick has neither a mechanical nor hydraulic physical linkage to any of the flight control actuators.  What we now take for granted, down to our cars,was pioneered back in the 70’s for space travel.  In sum, the space shuttles are flying baskets of complex interdependent computer systems, where everyone has a common purpose – tightly link all systems for safety and functionality.

If NASA has been obsessed with tightly integrating disparate computer systems, why then, is it common practice, when acquiring and merging distinctly different companies, that Post Merger Integration (PMI) of the underlying IT infrastructure is rarely addressed successfully?  Is it too expensive (not as expensive as when it fails)?  Is it too complicated?  Does the IT PMI process take too much time? Is IT a necessary evil which strategists don’t fully understand and therefore don’t wish to tackle?

My guess is that most of the financial calculations for M&A activity conveniently “forget” to include the many different types of ITPMI expenses required to make the new combined enterprise run smoothly and efficiently, ranging from picking which partner has, by functional area, best of breed systems, to adjusting rules engines, to consolidating forecasts and Financial and Statutory reporting. Most process and operations improvement consultants, as well as financially oriented executives have precious little real IT knowledge and experience at the Enterprise level, so perhaps it’s just a case of not knowing they need to focus really hard on this subject.  In this day when most revenue is somehow ebusiness or IT delivered or augmented, IT is a key value driver and not integrating properly is simply no longer an option. Excel and ‘sneaker-net’ are no longer viable forms of inter-company integration.

The next time you are considering any M&A investment, you would be wise to include this necessary expense. If the space shuttle can perform with revamped 1970’s technology, we can make our current technology sing off the same page.

Bruce Thom is the Western Region Business Development Director for Return on Efficiency, focusing on Hi-Tech.  He has over 30 years of selling to, and for, major software and hardware providers and sees companies from the unique perspective of an ‘inside-outsider’.  He can be reached at RoE’s website is

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: