This time last year I posted a blog about our trip
back to Silicon Valley where I played tour guide to my good friends, Dieter and
Chris Monch. Dieter and I had worked together back in Sydney, Australia, where
Dieter had been Managing Director of Nixdorf Computers. Way back in the early
1980s, Nixdorf Germany bought the company I worked for, The Computer Software
Company, renaming it Nixdorf Computer Software Company and leveraged the
product, EDOS (and EDOS/VS) together with the skilled team of systems
programmers to launch the Nixdorf line of IBM Plug Compatible Mainframes
(PCMs).
That’s right – PCMs from a manufacturer of
key-to-disk systems, ATMs and an integrated hardware / software small business
software package. Again, hard to imagine the thought process behind entering
the PCM business with the slowest (but least expensive) alternative to IBM’s
then 4331 / 4341 product line. But then again, Olivetti and Phillips were both
on a similar trajectory so Nixdorf didn’t want to be left out. Working with
Dieter on this new endeavor, in 1984 we managed to sell one more 8890 (as the
Nixdorf PCM product family was called) than Nixdorf US was able to sell in all
of North America. While there were celebrations back in Sydney I was left
dumbfounded and concerned – what was the future of Nixdorf ‘s PCMs if Nixdorf
US couldn’t sell many times more of them than we were able to do in such a
small marketplace?
The tour of Silicon Valley and then down the coast
to LA and across to Palm Dessert / Palm Springs was done in sunshine which isn’t
unusual for California, but even as I write this, the seventh snowfall of the
season has begun here in Colorado. Our orange Range Rover Evoque ragtop has
become our go-to vehicle, but unlike the week spent in California the top
remains firmly in place. I have posted a memory about this trip to Facebook but
what really sticks out in my mind is that reliving past glory days with my
former boss was a reminder that just when you think our IT industry has set a
course for us all to follow, something comes out of nowhere and we all have to
rethink our business models. It may all be sunshine and smiles in one place but
somewhere ahead of us, clouds are forming and they don’t look all that
friendly.
For the NonStop community, there have been numerous
solutions targeting the needs of enterprises running mission-critical
applications. These could be in finance, transport or manufacturing. I still
smile whenever I think of the story I wrote a number of years ago following an
exchange with HPE about a bakery in Japan that deployed NonStop. Included in my
white paper – HP
NonStop systems as you haven’t seen them before
that is still available on the HPE Tech Library
- the bakery had to ensure bread
(and I imagine, croissants) made it to shops up and down Japan, right on time.
No point in delivering these goods late at night as the needs of morning people
just had to be met. Mission critical? Absolutely! As for the white paper, if
you haven’t read it as yet you might be surprised to find how relevant the
storyline remains today.
However, it’s worth noting that NonStop is a lot
more than its hardware and it’s a lot more than the operating system. NonStop is
an integrated hardware / operating system / utilities, tools and languages that
collectively make up the powerful ecosystem that ensures that there is no
single point of failure. This ecosystem also delivers availability, scalability
and security unmatched by any other vendor with out-of-the box solutions. This
last point being absolutely the most important aspect of NonStop – you don’t
have to do a thing! Yes, you can create an environment that is similar to
NonStop using enablers like IBM’s mainframe Parallel Sysplex and you can even
pull together something that gets close to NonStop with Oracle clustering but
the resultant solution is only as good as the first release and can only be
supported with the retention of the original developers.
Too strong an observation? Once you get past your
first release of your custom continuous availability solution then the issue of
it being the sum of many parts, each in different stages of their life cycles
and tracking to different update release distributions, then you the user
become a product company necessitating the addition of staff that you find in
the bigger development houses – and paying them competitive salaries. Not a
long term winning solution by any measure you care to establish. However, it’s
very real and even in times where the mantra where good enough is well, simply
good enough, enterprises deserve much more from their vendors.
This is not the end of the story. What might be
driving success for NonStop in 2021? There is no question that there are challenges
lying ahead for the NonStop team and these challenges are happening on two
fronts. The first front just happens to be internal; NonStop needs to further
improve its visibility within HPE. There are positive signs coming from the new
leadership of NonStop’s organization who was positive about NonStop in recent
remarks to the NonStop community. This is in evidence too as there are equally
as positive signs “coming from the internal HPE sales side, which is actually a
good thing” I heard recently. And this is a good thing as we all know that
sales in the broader marketplace can only come as the sales force grow more
confident in the value proposition of NonStop. And we all know that “the
opportunities for NonStop lie outside the existing customer base as that market
is a lot bigger than the existing NonStop customer base.”
The second challenge NonStop faces is adding new
solutions. Shame on companies like SAP that simply don’t get it – in today’s
huge and fast growing eCommerce market, for a company as large as SAP to not
understand the value proposition of NonStop while continuing to accommodate failures
is yes, a real shame. SAP isn’t the only company that I have seen acting
arrogantly in this regard. Even as SAP has experienced many head-shaking
moments – think MillerCoors and Revlon / Elizabeth Arden – where a
fault-tolerant platform certainly wouldn’t have gone amiss – having had close
exposure to Epic at our local hospital I have to believe Healthcare solutions
need to target NonStop as well. And then there is transportation – what other
system can run container terminals 24 x 7 better than NonStop?
Again, many of these vendors do need to be ashamed
of themselves for not adding the option to run on NonStop a part of their
product portfolio. But there are answers
and they can be found in the ecosystem that makes up NonStop. Collectively,
they have made migration to NonStop a straight-forward proposition. There are
NonStop team members ready and willing to help out – the Advanced Technology
Center (ATC) being the primary go-to folks with help only a call away – as
there are numerous consulting vendors specializing in NonStop. As we were
pulling together the January issue of NonStop Insider we referenced once again
the post to HPE by Karen Copeland, Modernizing
the development world of NonStop applications where you will
read of the many tools, utilities and libraries that are now available to any
developer looking to support NonStop.
As the 1980s moved along, the world of IBM PCM
became fraught with lawsuits as first it was Hitachi than IBM they went after
and then Fujitsu. Ultimately, in winning some cases, IBM lost the battle and
even as Fujitsu won arbitration deciding that “Fujitsu has the right to sell
close imitations of key IBM software products, the die was already cast. Unix
was about to descend and alter the landscape irreversibly. Today, there is
little thought given to compatibility with one vendor or another as it’s all
about open source and just as quickly as Unix flourished, Linux cast a long
shadow over the industry whereby Unix has “left the building.”
There is one more answer that can be addressed and it's something that is really worth consideration. Driving our Range Rover convertible along the Pacific Coast Highway, we were reminded of how this car replaced a much smaller Mini Cooper S roadster. The enjoyment this Mini gave us led to us naming our Range Rover, “Mini on Steroids.” Each time you turn on the ignition, the Mini on Steroids message scrolls across the infotainment center. We cannot take for granted the ease though with which we moved from the Mini to the Range Rover nor the fact that given how both were ragtop convertibles, we picked up all the advantages of the Range Rover without giving up the flexibility this provided.
For many in the industry, the mere fact that NonStop has outlived so many platforms and remains unique in the way it addresses fault tolerance combined with the reality that applications developed for Linux can be readily (and somewhat painlessly) migrated to NonStop, can we take NonStop for granted any longer? Given the commitment to industry standards and open source that has been made by the NonStop team, would it be too much of a stretch then to consider NonStop, “Linux on Steroids?”
Comments