Wednesday, May 29, 2019

Legacy IT - a world apart from NonStop!

HPE is pleased to have migrated to a very modern NonStop at the heart of its data center – what about you?

I have no ambitions whatsoever to become a collector of antique cars. Try as I might, I can see nothing attractive about vehicles built 25 or 50 years ago even as I know so many of my friends hold dear to them the vehicles that they once envied in their youth but never could afford. We have a 2003 Corvette parked in the garage and with each passing year I wonder whether or not we have kept it for too long. It’s only fifteen plus years old but Margo and I are in no hurry to sell it as it is our only choice for track day fun outings. On the other hand, there is so much about this Corvette that even with the passage of just a few years, looks “old fashioned.” There is barely an electronic aid to be found and with the exception of traction control and ABS, it’s close to being considered an “analogue car”.
Perhaps one reason for keeping older cars with few electronic aids is that in the very near future they may disappear from the landscape entirely. You drive your car? You actually push on pedals and have a steering wheel? I can almost hear the shocked expressions coming from our grandchildren in a decade’s time when cars such as these are banned from public roads. Suddenly, not only is there a notion that motor vehicles will be looked on strangely, but their owners will likely be viewed as eccentric. Just as horses disappeared from our busy thoroughfares to become expensive hobbies, so too will cars – a legacy, if you like, of technology long forsaken in the name of progress!

When it comes to computing and what populates our data centers, I cannot recall ever reading as much as I have of late concerning legacy systems. I once worked alongside a colleague who owned a number of computer companies and one of them always brought a smile to my face – Archaic Computers. This was a forward thinking enterprise that bought up old mainframes, compressed them in order to fit as many as possible in a standard shipping container and then sent them to a company in Arizona. Gold and other precious metals were then extracted from the crushed remains such that it became very profitable for a short period of time until, that is and unfortunately for my colleague they ran out of old mainframes. At a time when we continue to discuss mining data for information “gold” it’s probably worth noting that data wasn’t the only gold locked inside mainframes. Gold just happened to be “the gold” back then!

In my discussions with industry and financial analysts the question always comes up about what constitutes a legacy system. And always the response is the same – it’s not that simple defining a legacy system. I know my colleagues working with IBM mainframes will be aghast at the news that general purpose computers, once dominated by mainframes, are today’s legacy systems. They have an old architecture, a convoluted operating stack, and a need for skillsets unnecessary elsewhere in IT. No matter how you slice and dice these systems and despite the virtualization provided, their costs outweigh any material value to an enterprise and, at best, they are marking-time until the applications they support find new homes.

A bit too tough on IBM? Perhaps; but when it comes to defining legacy there is much the business of computing has in common with the auto industry. In both industries, everyone will have an opinion but the opinion we all can relate to might very well be, “I may have difficulty describing legacy but I know a legacy system when I see one!” Sit inside a new car on any manufacturers’ showroom floor and almost immediately, the discussions center on the value proposition of the infotainment center – can you communicate with Apple CarPlay? So much for the old methods of choosing a car and yes, what lies under the hood? And yet, when it comes to modern cars it’s all about the graphics and interfaces and less about the basic engineering!

In computing terms, the word legacy is used to describe outdated or obsolete technology, equipment that is still being used by an individual or organization long after it should have been replaced. More noticeable still, “vendor or manufacturer support is not available for legacy systems and applications.”
This explanation came from an online dictionary I referenced where I also came across the following explanation of legacy, “In computing, a legacy system is an old method, technology, computer system, or application program, ‘of, relating to, or being a previous or outdated computersystem,’ yet still in use.” Yet another source I referenced, Techopedia explains Legacy System as a system not necessarily “defined by age. Legacy may refer to lack of vendor support or a system's incapacity to meet organizational requirements.”

In the case of an analogue car as described earlier in this post, there is value – there is both an emotional connection as well as the challenge that comes with physically managing the driving experience. However, when it comes to computing, apart from the emotional connection we may have developed for the very first computer we used – some colleagues have mementos, saved bits and pieces following dismantling, sitting on shelves in their offices - there is little that can be said about any potential enjoyment that comes with operating an old computer. Legacy computers are just that – reminders of once glory days that are best left for discussions at user events long after the effects of adult beverages have taken hold!

Techopedia, in its explanation of Legacy Systems, noted that, “
Legacy systems are high maintenance and may involve intricate patching and modifications. Porting techniques are often used for software adjustments or adaptation. Older hardware may require added compatibility layers to facilitate device functionality in incompatible environments.” On the other hand, Techopedia recognized that, “An organization might continue to use legacy systems for a wide range of reasons, such as the following:
  • "If it ain't broke, don't fix it!" The system might work adequately.
  • The system is complex, and documentation is poor. Simply defining scope can be difficult. 
  • A redesign is costly, due to complexity or monolithic architecture
In other words, and perhaps one of the best descriptions of legacy systems, Techopedia tells us that “A legacy system is a technology that is out of date to the point of representing an operational risk to a business.” Yes, it’s just plain old …

HPE NonStop systems never fitted in with the mainframe crowd. Indeed, those vendors championing mainframes back in the late ‘70s and on into the early ‘90s went to great lengths to push to one side NonStop systems whenever discussions about mainframes arose. Transaction processing “front end” computers were a specialty item and as such, lacked the wherewithal to do the full mix of work generally associated with mainframes.
In describing Fault Tolerance in the NonStop Cyclone System, writers Scott Chan and Bob Jardine began with, “The NonStop Cyclone system is a fault-tolerant multiprocessor mainframe designed for simultaneous transaction, query, and batch processing.”

Ouch – it would take a decade or more to move on from this even if it was all part of a positioning message to convince the global IT community that Tandem Computers was a viable systems’ vendor.
Sure, NonStop could do batch, after a fashion, but that wasn’t really part of what NonStop systems were designed to do – small message in, small response out, and a little bit of database access in between. “Do you have the $50 in your account that you plan to spend?”

What we have today with NonStop is far removed from legacy systems we know a lot about (when we see them). NonStop is not a mainframe and perhaps more importantly these days, NonStop is an active participant in HPE’s transformation to Hybrid IT strategy. From a company that now has a strong penchant for talking about software and software-defined-everything, NonStop can be licensed as software today to be run on hardware from server vendors apart from HPE. This is simply not possible with legacy systems of any kind that I am aware of. Indeed, the mark of a legacy system is that the system “is unable to scale to handle anticipated business volumes; has a limitation such as an inability to handle large numbers; technologies used by the system including business software, programming languages, operating system and hardware are no longer supported.”

At a time when the HPE NonStop organization is selecting partners to help enterprises with their migration and modernization projects as they embrace the new NonStop, it’s good to see that there is no references being made to the former family of NonStop – the NonStop i Systems – as being old or in any way, legacy. These NonStop i Systems could scale, handle large numbers and yes, supported modern languages, utilities and tools including a plethora of products and features created by the open source community. It’s always been my belief that the role NonStop has filled in the past has set up perfectly its role for the future. Availability will always be the gold standard of modern systems and with today's’ new NonStop there is gold aplenty to be had when it comes to meeting the business needs for true 24 x 7 sustained operation.

For many of us, there is a reticence to talk about legacy systems. There is almost a fear that even mentioning the word legacy might encourage those around us thinking of NonStop as part of a legacy world. Unfortunately, in our reticence to talk about NonStop and legacy is the very clear perception that we are being evasive – we don’t talk about it so it must be true: NonStop is part of the legacy world! However, NonStop was never part of that legacy world and today, remaining modern, we see NonStop – a vital cog in the transformation to Hybrid IT machine – offering even more options to better capitalize on the growing world of transactions, be they originated in traditional or virtualized environments.

Whereas HPE understands it is very fortunate to be the home of NonStop – is your enterprise, on the other hand, pleased to be moving NonStop into their home? After all, there really isn’t another viable alternative to NonStop when you get right down to it; yes, it is still very much a case that, in today’s world of computing, a very strong argument can be made that #NonStopRocks !!!

Tuesday, May 21, 2019

ETBC, Edinburgh – as successful as we expected it to be!

Well done, BITUG committee! Edinburgh, Scotland didn’t disappoint and the level of participation was tremendous with all keynote sessions very well attended …

We left for Edinburgh, Scotland, as planned. We had dinner at the airport, caught a flight to London where we enjoyed lunch at Euston Station before taking the train to Edinburgh. All rather straight forward with the only hitch being that we should have caught the train from Kings Cross as our Virgin “express” stopped almost everywhere on its swing through the west of England and we arrived at our hotel some six hours later. Saw plenty of green hills and lots of sheep, mind you, and the weather was way warmer than we had anticipated. Yes, last Wednesday Scotland enjoyed warmer days than residents in Spain and even LA experienced.

There will be much more coverage of the social side of the event and the experiences we enjoyed in the company of other attendees will be covered in an upcoming post to our social blog, Buckle-Up, but suffice to say, Margo and I enjoyed the time in Edinburgh. As an aside our first trip to Edinburgh and to this same Sheraton Hotel was back in the spring of 1999. It was the time when there was a Spring European ITUG event each year and waiting for a lift to take Margo and me down to the exhibit hall, we met Robin Gilchrist and Tony Bond – Robin was then ITUG Chair and Tony was the Vice Chair. Little did I know that this brief encounter would be life changing after a fashion, as subsequent conversations with Tony led me to run for a director’s position on the 2000 ITUG board!  

I mention this in passing not only because of the impact it had on many decisions I made in the years that followed but as I took to the stage for my own presentation, I looked out at Yogesh Teli and Margo Holen who both joined the ITUG board at different times and the thought I had was that well, we almost had a quorum so what should we be discussing this time around? I have always enjoyed the company of Yogesh and we travelled the world together for a time catching up with each other in South Africa, Singapore, Australia and yes, the U.S. Yogesh immediately preceded me as the ITUG Chairman (in 2003) and it was being his Vice Chair where I learnt so much about what to expect as an ITUG Chair. So Edinburgh holds fond memories for me and seeing Yogesh and Margo seated in the front made me feel quite at home as I began my presentation.

Last year I thought the turn out for the GTUG hosted pan-European event in Leipzig was outstanding but seeing 230 plus folks make the journey to Edinburgh impressed me just as much! To the BITUG committee, I tip my hat as they did a wonderful job in making this event work the way it did – from a traditional beer bust to kick off the event Monday night to the gala evening out at the Hub alongside Edinburgh’s famous castle to where a program was created that easily blended the news from HPE and the NonStop team with numerous customer and vendor presentations there were plenty of informative sessions to choose from. After polling just a handful of the exhibits, the vendors I spoke to were somewhat exuberant when it came to relating the much-improved traffic flow to their stations than previously experienced.

As Tim Dunne of NTI said, “We had major financial institutions stopping almost constantly by and asking us about how we can address their business needs. It was hard not to be impressed with just how many customers had turned up for this event.” OmniPayments’ Craig Lawrance added, “I have had better conversations this year than in any other year that are already leading to more opportunities than ever before!” With the BITUG Chair being TCM’s Collin Yates, the TCM table was always busy as was that of ETI where I had a tough time getting the attention of ETI COO, Sylvain T├ętreault. Once again and somewhat predictably, put the coffee and then the lunchtime food in the middle of the vendors’ stands and everyone is happy!

I guess by now everyone has heard the news that the end of sales of Itanium systems will now be coming forward for everyone in EMEA – rather than ending July, 2020 they will end in January 2020. The reason for the change was pretty obvious, according to NonStop Enterprise Division boss, Neil Davis, “It’s ServerNet and the lack of availability of ServerNet related hardware,” that has influenced this decision. And perhaps, after talking to NonStop customer and vendors, this is a moot point as migrations to NonStop X have really begun in earnest of late and that is a very encouraging sign for the NonStop community as a whole. Initially there had been some pushback by the NonStop solutions vendors but HPE has done a terrific job in getting resolutions to NonStop customers’ potential issues to do with migrations.

As for the question of whether or not one vendor or the other was going to migrate or even an issue about the NonStop team supporting mission-critical middleware many NonStop users depended upon, today we know BASE24 will work as will Lusis, even as we have seen Oracle support for GoldenGate extended to include NonStop X. Issues surrounding support of IBM’s MQ are being addressed and as for the NonStop SNA and X.25 products, resolution has arrived. If you depend on SNA or X25 and have relied upon a SNAX or X25AM product, then going forward NonStop has elected to add the Infrasoft’s uLinga product to the HPE NonStop price book.

In a deal worked out with comforte, the sales partner of Infrasoft responsible for sales and support of uLinga, the NonStop team has made a somewhat momentous decision. Both Margo and I were heavily involved in SNAX and then later ICE that has seen us today committed to the success of uLinga and while it took almost a decade to achieve, we always knew there were better solutions available today than the legacy SNAX and ICE products. With uLinga you will be getting a lot more than just a replacement for either SNA or X.25 product offerings but access to features that better integrate NonStop TS/MP applications with both CICS and IMS where there isn’t anything in between but pure IP!

However, these updates were really only a starting point for the event in Edinburgh. ETBC was full of announcements as one after the other vendors talked up their plans for virtualized NonStop, consumption-based pricing and for many, their support of NSaaS – yes, NonStop-as-a-Service.  There are many product offerings from the NonStop community that lend themselves to being accessed out of a cloud, whether it’s a public cloud or more likely than not, a vendor supported private cloud (and there are a number of NonStop vendors well advanced in rolling out their own global clouds all based on NonStop – again, OmniPayments and NTI being two that come to mind immediately), so be prepared to find a number of new and exciting ways to consume your favorite product.

Key candidates? It you are after monitoring of you system and application then I suspect there will be cloud offerings. If your need is for real time analytics to be performed on select data then that too will likely be supported out of a cloud. And what of data replication, file movement and distribution, and yes, products that ease your transformation to hybrid IT? For many of the products I can think of there is almost an equal number of ways accessing them so running these from out of a cloud could easily apply. When you consider that today, NonStop has become a collection of VMs (versus real hardware machines), then the possibilities will be endless and it’s really all confirmation that NonStop is modern and more than an appropriate choice for any business looking at the bigger picture of digital transformation.

If events like ETBC tell us anything at all then it is that HPE continues to invest in NonStop and that NonStop is increasingly turning to its partners. Indeed, as IT comes to terms with Digital Transformation and the world of Hybrid IT – it’s becoming less about partnerships and more about ecosystems. Winning vendors are building out ecosystems that include many vendors who in turn are cooperating with other vendors. Point is bigger vendors will be unfolding umbrellas under which you will find numerous vendors all working towards the same goal – to provide better customer experiences, whether those customers are end users, operators, data scientists and more.

The signs are already there telling us that vendors are talking to vendors even as HPE NonStop is providing a helping hand all of which is to say – success with the core NonStop business as in evidence with the data provided by HPE’s Neil Davis is proving more than enough incentive for vendors to begin partnering-up to create their own ecosystems beneficial to us all!  

This year’s ETBC was a success even as NonStop is beginning to enjoy success in its own right. There will be a lot more events held in the coming months none bigger perhaps than HPE Discover where there is always something said about NonStop but for now, it’s good to be able to reflect on all that we heard in Edinburgh. As for next year all the signs are pointing to a big event again, but this time in Berlin, Germany. Until then, all we can add is our congratulation and thanks to the BITUG committee for putting on one heck of a show! Well done, team!

Thursday, May 9, 2019

Another weekend and another European escapade!

Has a renewed sense of urgency taken hold inside HPE that is (positively) influencing the HPE NonStop team? Is HPE looking like a start-up, once again? Whatever you call it, a lot of positive news is coming from both of them and it's not just their new digs …

For a company with staff that doesn’t fly all that often, the Pyalla Technologies team – that’s Margo and me – will be heading to the airport late Saturday to begin our journey to Edinburgh, Scotland. Already this year we have flown back from Australia in January before escaping to Munich, Germany, in March for a couple of adult refreshments together with a couple of client meetings. Turns out this return trip to Europe in May, for the European (NonStop) Technical Boot Camp (ETBC) will be followed with a visit to HPE Discover in June. We have even managed to squeeze in one more trip to participate in N2TUG on June 27.

It also turns out that there is a significant uptick of interest in Regional User Group (RUG) events, no matter where they are being held and we want to see this one for ourselves. As for the reason, it’s not just about checking out the attendance but where else can you hear about product investments being made by HPE in NonStop? A recent conversation with a colleague of mine reminded me of one key ingredient that is making RUG events popular – yes, it’s the only place to hear the latest news about NonStop. Forget social media, he suggested much to my chagrin, you just have to make your way to the nearest RUG event to fully understand all that is taking place with the NonStop product line.

While I am expecting to be surprised, perhaps more than once, during this year’s ETBC event even as I think I have a good understanding of what priorities NonStop development has set for itself, there is bound to be something new announced. Just as we have heard much about how NonStop developers are improving the value proposition of NonStop they clearly haven’t eased up on developing new features. The probability that NonStop will attract more customers and development partners is high and quite honestly, it hasn’t been business as normal inside HPE for some time.

Ever since HPE CEO Antonio Neri ascended to the leadership role, he has brought attention to where HPE is headed and it’s mostly about providing realizable value up and down the HPE product line. With his Next initiative (to re-architect the company), Neri has been extremely vocal, stating that "At the core of this is not just cost savings. It is all about simplification, innovation and execution."

So, is HPE looking more and more like a start-up? With as much talk as there is of late about innovation, and the need to innovate “at the speed of business” is HPE becoming an entirely new company? Has the recent move to new “digs” in San Jose all part of a plan to refashion HPE to become more nimble? The essence of Neri's Next transformation, according to a May 22, 2018 update in CRN, HPE CEO Neri: Next Initiative Is The Ultimate 'Competitive Advantage' In Fast-Moving Market, “is a return to an ‘innovators at heart culture.’ HPE has acquired assets and innovated around those core competencies in a way that gives the company a deeper and wider story. It is not piece parts.”

One of the key attributes that distinguishes a start-up from more established enterprises simply pursuing business as usual is the observable sense of urgency. The need to get things done with products introduced into the marketplace quickly! According to a McKinsey & Company strategy paper published in May of 2019
How to move fast: Innovation at speed and scale, “The biggest thing that makes most start-ups move really quickly is urgency. They are running on the amount of capital and the amount of funding they have at any given moment. And if they cannot get to a next milestone, whether that is actually getting to enough sales to start self-sustaining the enterprise or get to a next milestone with a venture funder, they’re done.”

Ouch! Clearly, HPE is not quite as pressed to succeed as today’s start-ups when it comes to capital. Or, are they?

When you look at the NonStop organization today it is a lot smaller than at any other time in its history. And yet we have all seen x86, InfiniBand and RDMA over Converged Ethernet (RoCE) followed by Oracle compatibility (for NS SQL), virtualized NonStop, capacity on demand and much more – a sense of urgency? You had better believe it! HPE talks a lot about the core and the edge but how many of us have truly caught the message from HPE of late.

"The digital transformation really starts at the edge. Sixty-plus percent of the data is generated at the edge. Two years from now, we are going to have twice the amount of data we generated in human history,” HPE told CRN reporter, Steven Burke, last year. “Analytics and AI are required at the edge. The compute capacity at the edge in the form of cloud architecture is going to be a significant opportunity for us. That is why I am very bullish about the future of HPE."

You may not view HPE as being in start-up mode just as you may not be thinking of NonStop being new to the market. However, there is no denying that NonStop and indeed HPE has stepped up the pace of innovation. At one point, McKinsey and Company partner Stacey Haas notes in this month’s strategy paper how, “It’s clear the many layers that larger companies have can blur the focus and urgency around decision making, and even make accountability diffuse.” However, under Neri this doesn’t seem to be the case and to that point; let’s just take note that HPE today is “only” a $30+billion company – can we even consider HPE as being big by IT industry standards?

In other words, this appearance of HPE as being in start-up mode and indeed, as is the NonStop group, is clearly a reflection of a rush to innovate together with a rush to get to the next milestone. And the milestones just keep on coming. And where is the NonStop community getting to hear about all of this? The “inside skinny” is one of the true benefits from attending RUG events and with the European event in Edinburgh – the ETBC, hosted by BITUG – about to take place, there is an expectation that we haven’t heard about all the plans for NonStop. There is more news coming so I am led to believe.

“The compute capacity at the edge in the form of cloud architecture,” intrigues me as it is generating renewed interest within many of the NonStop vendors with whom I work. From my perspective, when it comes to the enterprise this is a reference to private clouds more so than public clouds. It’s where transactions take place and it will be where “analytics and AI are required.” Sounds a lot like shortly we will see virtualized NonStop (vNS) being put to good use. Perhaps there will even be members of the NonStop vendor community stepping up with utilities and tools to help make this happen.

The NonStop team has a lot of projects that they categorize as being “under investigation.” Take vNS on HPE Synergy, for instance, as this is something that continues to intrigue me as well even as I have evangelized for some time that this looks to be taking a step in the right direction. Then there is all the effort being made to bring even more open source to NonStop which is a meaningful way to reinforce the overall open message of NonStop.

On the subject of open and just to reinforce the importance of RUG events, if you haven’t seen the promotion for the upcoming June RUG event, N2TUG then you may have missed reading about the keynote presentation by HPE’s Meg Watson of the ATC, who will be sharing how “industry-standard cross-platform tools, such as Jenkins, Git, Eclipse, and Ansible can be leveraged by your organization to help you meet the quickening demands of your business without missing a beat!”

Once again this, and much more, for most of us will only ever be explained in detail at a RUG event – getting the picture now? If you follow LinkedIn or Twitter you may get a glimpse of the urgency evident in HPE’s programs even if you don’t see specific references to NonStop. However, to hear more about the specifics you just have to make it to your local RUG events – and I cannot stress too much the importance of doing so!

HPE is moving fast and is innovating at the speed of business but you will likely miss hearing about any of this if you aren’t attending pivotal events and even if it means catching a plane, being present for any RUG event is certainly going to help you to better understand where NonStop is headed.

NonStop, whether traditional systems, virtualized systems on HPE hardware or on your own hardware or within your own private cloud – you now have options. And whether you see the future of NonStop remaining in the core or gravitating to the edge, the most important thing to remember is no matter the decisions you may make, the big news is that yes, the news about the future of NonStop is simply that for NonStop, there is a future!

Looks can be deceiving! HPE NonStop; when being the best still matters!

For the NonStop community, we know what looks good may not only be deceptive but borderline dangerous; mission critical applications are bes...