Monday, June 29, 2020

Changing our perspective; what develops will be good!

HPE Discover is always an occasion that the NonStop community looks forward to and with the event going virtual, HPE met our expectations …

One of the benefits of living in Colorado is that the weather is truly a hit or miss circumstance. For the past week we have ventured out into morning sunshine only to be pelted a few hours later by heavy rain and yes, hail. Whether it’s ducking back into stores that happen to be open or jumping into our vehicles it doesn’t really make a difference. Living in Colorado, alongside the Rocky Mountains, means making adjustments quickly in response to changing circumstances no matter what plans were in place. And yet, after checking with the locals, apparently Colorado enjoys more sunny days than just about any place else in USA.

Among the many changes of circumstance that the global pandemic has brought about is that for Information Technologists the big-tent marketing events put on by major vendors have undergone change that will likely live on well after any sense of normalcy returns. Nowhere was this more apparent than with HPE’s annual trek to Las Vegas for HPE Discover. It may be June and the weather may be openly hostile but every year for as long as Margo and I have been associated with HPE, there has been an event in Las Vegas. And we have travelled to each and every one of them beginning with HPTF 2008!

For 2020, with the global pandemic still in full swing, HPE Discover 2020 became a virtual event and I have to say the effort made by HPE to ensure a sense of community paid off handsomely. From the early press, analyst and influencer briefings that began early Monday on through to numerous entertaining sessions with HPE executives along with industry and sporting personalities were very well prepared and interesting – I even had the chance to ask questions of Susie Wolff and Felipe Massa of the Formula E Venturi team, of which HPE is one of the sponsors.

I have to admit I went into the week a little wary of how it would be received. I also have to admit that I wasn’t all that upbeat about this event’s prospects of attracting a crowd. And was I wrong. Numbers were clearly up across the board – if you hadn’t registered for the hands-on labs that were being held that week then you were plain out of luck. They were essentially oversubscribed even as I am still unsure what a virtual hands-on really means other than what I did hear back from those who managed to participate. There were dedicated systems set up for the purpose and these virtual attendees got real world experience with newly announced software.

To say I had a change in perspective sums up my own experience. Once again the guest of HPE as a member of its influencer program, HPE manager Laura Mackay and her team went to great lengths to keep us all engaged and throughout the event there were numerous occasions where Laura simply invited us to log onto Zoom for “Bloom Networking Events” where there was no agenda other than each of us updating the others on what we were doing and then joining in on a free flowing exchange about our experiences to date. And yes, these sessions were held in good spirits and were fun … and we all enjoyed a kind of informal social even as some of our European colleagues were participating at very unfriendly hours.

We each received SWAGs from Laura and as in former times we each ripped into the package as it landed on our doorstep. Sweat pants, coffee mugs, a HPE custom Rubix Cube, Apple earphones, and more … but again, it was the opportunity to hear directly from the HPE team that held our attention. It had only been a few days earlier that HPE CEO Antonio Neri advised the world that he had tested positive for COVID-19 and yet, here we were, logged into the HPE network, listening to Antonio early Monday morning as he held a press and analyst pre-briefing on the as-a-Service and HPE GreenLake news that HPE was releasing at Discover Virtual Experience.

All under NDA, of course, until early morning the next day when Antonio gave his keynote address. I have to admit there was more than one of us caught by surprise over the steps HPE was taking to truly differentiate itself from the competition. HPE is now one year into a three year program whereby it is focusing all of its energies on becoming the edge-to-cloud platform “as-a-Service” company. “Edge to Cloud, delivered as a service is the right solution for today,” said Antonio. “Furthermore, we are entering the age of insight; more informed doesn’t mean more intelligent. The edge is where we live and work (and now we are delivering) on the next wave – bringing the cloud experience deep inside the core; to the data center.”

In a separate presentation, Antonio said:

“I believe we are nearing the end of the information era, which focused on generating and collecting massive amounts of data—data that could not be brought together to create timely insights and actions to change our future.

“The next decade must be about insights and discoveries that are shared and elevate the greater well-being of every human being on this planet. 

“The age of insight requires new principles and priorities for digital transformation. The focus is to build an edge to cloud platform that connects, protects, analyzes and acts on all of your data and brings agility to your apps to unlock your enterprise’s full potential.”

Pointnext; GreenLake, Ezmeral – what do they all mean? And what exactly is the Edge? As HPE provided more information about its vision it was clear that the Edge was much more than a sensor or terminal. For many enterprises, the trend was towards populating mini data centers as it was becoming more important to process data as it was being created. The Edge is the manifestation of intelligence being moved closer to the source. When announced, Pointnext was billed as being the HPE technology services organization in support of Hybrid IT but as of May 2020 its leader, Pradeep Kumar, reports directly to Neri. As HPE accelerates it’s move to as-a-Service, it is Pointnext where the skillsets needed to oversee as-a-Serivce deployment reside.

GreenLake? It too can be viewed as a service that brings with it products heavily oriented towards software. If you want to run your IT as-a-Service, who is tracking your usage? GreenLake Central is a hub (supporting a dashboard) where the data is located and where HPE looks for updates on usage – think Pointnext as financing your IT resources and GreenLake as collecting data on how those resources are being used. All under the general heading of only having to pay for what you use …

Ezmeral is a branding exercise that brings together all the HPE software needed to ensure that you not only can deploy your IT as-a-Service but where you can bring the cloud experience you see today on offer by the major cloud service providers, in-house. One set of tools and utilities with the added flexibility that from a dashboard you can assess the cost benefits of running an application in-house or in a cloud. Flexibility in ways unimagined only a short time ago and yet, with only 30% of applications and data running in clouds today, this will help modernize and indeed integrate what remains in-house. The easy applications were moved quickly to the cloud, but now those hard to move applications can benefit from cloud-like oversight.

CRN reporter Steven Burke saw it and following HPE CTO Kumar Sreekani covering the introduction of GreenLake and introducing the new HPE software brand, Ezmeral, Steven concluded that as of now, HPE Ezmeral Is ‘Undisputed Leader’ In AI Battle With AWS, VMware, Red Hat: Kumar Sreekanti If you read just one article published following HPE Discover 2020 then this might be the one you want to choose – just follow the hyperlink above.

Of importance for the NonStop community came much later statements by Antonio with none more important than the following: 

“It is not just about GreenLake because we have a transactional business which is very large and we are pivoting to as-a-service, which is the long-term future. We need to be able to do both and be able to give partners the flexibility to come along.”

Delivering Edge to Cloud as a service is the vision but it’s also complementary to what HPE delivers today to its enterprise users. The subtle twist however is that even with its transactional business HPE will not be excluding NonStop from GreenLake. It may not seem apparent at the moment even as there is work to be done by the NonStop team to complete the journey to “as-a-Service,” but it’s safe to say that this journey for NonStop is being investigated even now.

Perhaps it was in the explanation of the goal for Ezmeral given by CTO Kumar that clarifies at least one future for NonStop:

“While transactional (business) remains popular, we have to change customer experience and engagement as it really is all about the end-to-end experience.”

If you aren’t sure about what this implies then it should become clear in the coming months – NonStop will be as easy to configure, deploy and operate as any other cloud service offering today with enterprises able to readily glide between platforms without any loss of the cloud experience. Not today, as there’s much to do. This may take a while to sink in but NonStop is a part of this journey and as such, will be subject to additional development work that brings it into this new age of insight. After all, NonStop creates data and it is fresh data created in real time that has the most to offer enterprises poised on making insightful decisions.  

Circumstances just like the weather can change rapidly and for vendors like HPE the future is changing rapidly. Very soon HPE will become more like a software and services vendor, with hardware lessening in priority. NonStop customers will still be able to purchase traditional NonStop systems but looming large on the horizon is a completely new way to view NonStop. And the good news that continues to develop? Whether you deploy NonStop out at the edge, close to the source of transactions or integrate at the core supporting database services, there are always benefits to be had with a solution that never fails. 

Friday, June 19, 2020

The blueprint to optimum data integration; we sure could do with some help, right?

Crossing the lines, stepping outside the box and even lining up incorrectly are all familiar warnings; for NonStop architects, lines on whiteboards or napkins oftentimes convey complexity that needs expertise to address …

I have been involved with the building of three homes. With each home, the engagement differed as the starting place varied, but there was always a blueprint on hand. In the southern suburbs of Sydney, Lilli Pilli, the home was already standing and it was down to details. However, I did get to finish out the lower level, lay the bricks for the backyard retaining pool and come up with the design of the swimming pool. 

As for my home in Niwot, North East of Boulder, I started with a blank piece of paper and took many months before settling on the final blueprint of my dream three-level home. Every detail was covered in those blueprints when breaking ground commenced, but even so, when it came time to move in, there were additional lines drawn on the original plans for each floor that was built.

My current home in Windsor, a little to the East of Ft Collins, was yet again another instance of a home already under construction, but even so, minor details were only finalized as the house was being framed and as for the lower level, it was all my own design, although loosely based on what the builder had originally planned. Each time I built a home I gained experience and you might even say third time, lucky.

When it comes to IT, as business needs evolve, there are always plans being drawn up for the next big thing! As excited as we may be, walking up to the whiteboard to take in the scope of the project, experienced IT professionals will know that as the project takes shapes, additional lines will appear magically, almost out of nowhere, on the whiteboard. Even so, fluidity is always present whenever major projects are initiated.

I was reminded of this as I prepared an article for the June issue of NonStop Insider. Those lines, as it just so happens, are oftentimes the most complex of functionality to properly implement. It’s not just that they represent the connectivity, but more importantly today are deserving of a second look as they can mask the complexity of what is being contemplated. 

As one vendor said to me just recently, “Complexity is in the connections. Those are the lines on the whiteboard ‘architecture’ filling the void between components; between NonStop and data targets for instance, where data created on NonStop is needed in volume and on time, all the time. Interruptions will have significant trickle-on impact where the consequences of lines being broken might prove fatal for the corporation.”

As a community, those associated with NonStop have really taken to the observation by Jeff Kyle, VP Mission Critical Systems at HPE, when he said, “NonStop creates the data.” That’s right; data is created on NonStop and as such, represents the freshest and most relevant data that business will ever ingest. Fresh data representing changing market conditions is more valuable than any service provided or any commodity manufactured. Without fresh data, business will always be playing catch up and be in jeopardy of losing out to someone new to the market that is better positioned to meet the changing market conditions.

We often talk about improving business outcomes but this is not something that can occur if we are only factoring into our decision data that was created yesterday. It’s not good enough to be working with data that is that stale. NonStop systems improve business outcomes as the insight they can provide is immediate and sometimes, we don’t do ourselves any good by being quiet about this when we walk up to those whiteboards and look at the lines connecting NonStop to other systems within the business.

There are lines drawn that represent pathways to backup sites necessary for meaningful pursuit of business continuity. Disasters will continue to happen and it’s critical for business to have those additional sites primed and ready to take over in an instant. But those lines may not go to a physical data center as they may be funneling data created on NonStop to the cloud. 

Cloud service providers have made great strides in creating hierarchies of storage such that it might prove more economical to move data into the cloud than to a traditional off-site secondary or tertiary data center. However, business continuity represents only one type of line, as increasingly today there are lines connecting NonStop with data warehouses, data lakes, and more where the thickness of the line drawn on the whiteboard may indeed be quite misleading.

It’s not just a communications link but rather it may be a key component of a powerful data integration solution. And this is where the complexity can soon overwhelm less experienced IT professionals and call into question the skillsets you might be able to leverage from your favorite consultancy. There are reasons for blueprints and whiteboard diagrams and it’s all quite simple, really. There has to be defined beginnings and endings just as there are places where other interested parties have to be informed.

These could include local and regional authorities as well as internal audit, security and yes, operations management – how do you expect us to run this thing if you didn’t let us have a peak before you started the project? We could have told you that the data already exists and you are now duplicating processes – if only you had asked! When you build a house then gaining a certificate of occupancy is a major milestone. But when you develop a new application and with in-house QA behind you, it’s a successful Beta / Proof of Concept / Pilot that really informs you that all the decisions made during the life of the project were the correct decisions.

Returning to those lines on the whiteboard, one more thought came to mind as I read the Striim submission to the June issue of NonStop Insider, NonStop users rely on Striim to connect to the outside world. Azure cloud service? Yes! Kafka messaging? Yes! Data may be created on NonStop and data may need to be moved, transformed, enriched and much more, but the data created on NonStop first target destination is the NonStop’s own data base and in most cases this just happens to be NonStop SQL/MX.

As for Azure or Google or for that matter, AWS together with NonStop SQL (and the lines we draw connecting the two), such pairing is becoming the new normal for connectivity. “Clouds are capturing the enterprise IT marketplace and in doing so, they are appearing on radar screens where NonStop has already been deployed; and we already have transactions creating data on NonStop capitalizing on Striim to pipe data to Google Cloud,” said Striim cofounder, Sami Akbay.

“Databases appearing in public clouds may look attractive but what are the steps required to ensure NonStop and the cloud service providers offering you selected happily coexist without degrading the availability attributes of NonStop?” When was the last time you stepped back from the whiteboard and had a good look at that rectangle you had drawn representing a NonStop system? And where did your thoughts take you when you identified the need for a data streaming platform?

As a general rule, it is within IT where we collect data and we deliver data. The locality of data created on NonStop remaining on NonStop is very important. Not just for the integrity of the transactions that might follow or the reduced latency from not having to navigate off-platform, but for data’s own protection. Even as we shy away from attracting too much attention to the topic but data is safest when it’s on NonStop and should anything happen to data once it is passed to other systems, we can always rely on NonStop to provide that one instance of truth.

The NonStop community is well served by vendors providing products that connect disparate systems and move data – NTI, ETI, Gravic, TSI and Striim. They all look to support those lines of complexity when given the opportunity and they all do a terrific job in this regard. They all move data and they all can make getting data off of NonStop a much easier task to accomplish. All perform their tasks professionally and they all are experts in fields related to the segments of the market where they operate. 

It was from actions carried out by these vendors that the first moves to integrate data occurred and lest we forget, NonStop vendors were supporting heterogeneous mixes of data base management systems well before the rest of the industry began thinking about life with more than one data base solution.

These lines are complex and they imply so much. Next time you turn to look once again at your whiteboard don’t be surprised at just how central a role NonStop is playing in the support of that new application. No server matches the uptime or the scalability of NonStop. Surprisingly, no server can match the entry price of a new fully functioning NonStop system – have you considered that aspect of “going NonStop” of late? 

You may want to take another look as you will be surprised to find just how competitive the new NonStop X family of converged systems has become. They may be only lines and rectangles but ultimately, it’s your business being blueprinted and in a world embracing change at the rate it is occurring today, can you afford to terminate any of those lines in anything other than NonStop?

Wednesday, June 10, 2020

Outages, Yes! But with NonStop, No!

In times when it seems defining what’s coming next has become a sport, it’s even more important to stress what NonStop means for uptime whether total outages or simply brownouts!

Stepping into the garage and checking on our hybrid powered vehicle, it’s always good to see that it is fully charged and ready to go. Margo and I are now onto our second plug-in hybrid vehicle and it’s encouraging to see the range of our second generation extended beyond the original 12 to 15 miles to where it’s now more than 35 miles. Doesn’t sound like a lot but in the world of plug in hybrids, relying solely on electric power isn’t what it’s all about – electric power augments a traditional combustion engine thereby considerably increasing the savings in fuel costs. We have enjoyed fuel consumption dropping to where on one occasion we eked out almost 80 mpg on a trip from Windsor, CO, to Dallas, TX.

Seeing the vehicle fully charged is a not too subtle reminder that I too am fully charged. I’ve never been one for the glass half empty expectation. For me, it’s always been about expectations of more to follow. By this I mean, as a community NonStop users and vendors have known for a very long time that the NonStop systems provide an unfair advantage to those who implement mission critical applications on NonStop. Unfair in the sense that there’s really nothing more to do to ensure maximum uptime whether planned as in an upgrade to a key piece of software or even the hardware itself, or unplanned. NonStop, doesn’t stop!

There has been a lot written about the “New Normal” or perhaps the “New Abnormal.” There’s even an article I have read that discussed the “Not! Normal.” At this time I have written posts and articles on all three of these declarations so it’s not something I want to revisit at this time. However, it is worth noting that in continuing to qualify normal, it is with considerable regret that there’s little chance that we will see a return to the old ways, whatever we want to call normalcy, we all enjoyed just a year or so ago. Whatever we will face in the near future will be very different from those times now receding quickly from view and becoming nothing more than a page in our history books.

It is worth noting too that business has become even more sensitive to outages than in the past and that there is just as much written about brownouts as there is about outages. In this context, in case you missed the inclusion of brownouts in assessing true availability, it is a reference to the inability of the system and the application to adequately handle unanticipated transaction spikes. SLAs are typically written around an expectation of a certain range of acceptable response times but when subject to these unexpected spikes, response times’ blow out to where for all sakes and purposes, the system and application look like they are no longer working.

All of which is to say that these systems cannot scale-out without the user taking a hit. Scalability is once again on the radar screen for most companies and the inability of their systems to scale-out has suddenly become an issue. If you cannot scale out to handle spikes then you degrade your recorded levels of uptime. A response time of thirty minutes truly is akin to a system outage and is beginning to be reported as such. And it’s being recorded as part of the metrics included in modelling of unplanned downtime.

It is not every day that I reference IBM funded research in a post for the NonStop community. In this case, though, it seems appropriate. IBM funded Forrester to survey business to better understand “The Real Costs of Planned and Unplanned Downtime.” It was commissioned in August 2019 so it’s a pretty accurate reflection of what is now happening. The tag line that followed this heading simple read, “accelerate recovery with new technologies.” No reference to fault tolerance or even continuous availability as we know it within the NonStop community, but all the same, it’s not a bad attempt to highlight what ails business today and for a quick summary of key findings, the chart Forrester provides is worth taking a look at: 

How about that opening at the top of the key findings? Planned and unplanned downtime is common within organizations. Really? Outages just have to be accepted as part of the new normal for data center operations. Surely not! Now there’s an observation that many in the NonStop community will definitely treat as the New Abnormal. With NonStop it’s as if the rest of the world, IBM included, still hasn’t read the memo – when it comes to running mission critical applications, choosing a fault tolerant system like NonStop means you don’t have to be a light sleeper always having one ear listening for that dreaded call-out phone call.

But then again, this isn’t normal at all for the NonStop community and it’s worth stressing. NonStop is more than just fancy redundant hardware. It is the collection of hardware, software stack and the application all working together that ensures NonStop alone satisfies the IDC AL4 criteria whereby there is no observable outage for any user. None! NonStop truly does deliver zero downtime and just as importantly, with its ability to scale-out to better handle unexpected transaction volume spikes, imperceptible brownouts. So, when Forrester asked the question, “If your organization experienced close to zero downtimes, what technology benefits would you expect to see?” there were some interesting responses.

“When asked about the benefits of experiencing close to zero downtime, 50% of IT professionals indicate that it would allow them to run maintenance more frequently, addressing a top challenge for planned downtime. Forty-six percent of IT leaders indicate that less downtime would mean faster data recovery, addressing one of the top challenges of unplanned downtime.” 

With no indication by Forrester that this was all doable today with NonStop, it is so humbling for the NonStop community that such big percentages of those being interviewed were already dealing with serious issues arising from their inability to deploy systems and applications without facing some serious downtime. Not being able to run maintenance online while all else continues to be running is definitely not the normal for NonStop customers.

As Forrester concludes its findings it provides in its summary something NonStop users are all familiar with even as it isn’t something that concerns them. However, it’s something they need to be much more aware of from the standpoint that it’s a big item to champion within their IT organizations. Forrester’s findings? How about:

“Unplanned downtime is unpredictable and it creates extra costs for organizations. From a technical perspective, IT leaders are concerned with the data recovery process during unplanned downtime. From a business perspective, IT leaders are concerned with the overall loss of revenue caused by unplanned downtime.”

Phew! As a community, NonStop users certainly are dodging a lot of bullets here. We cannot stress this enough. There’s no better solution for accelerating recovery than NonStop and why? It’s fault tolerant through and through and it’s happening transparently to the end user. They see nothing and when they see nothing, there’s none of those dreaded nightly calls to senior IT management. There is always the consideration of business continuity planning involving secondary and even tertiary sites but that too is so much easier to accomplish with NonStop with the assistance of knowledgeable partners with a plethora of product offerings.

As Tim Dunne, NTI’s Senior Vice President Worldwide Sales, said in NTI’s article in the May, 2020 issue of NonStop Insider, NTI watching what’s changed: the need for social distancing and what hasn’t changed: the need for business continuity -

“The focus on Business Continuity remains as strong as ever and for good reason; no business wants to be offline as the world of business becomes a fully online business. As an industry we can debate whether it’s happening in the data center, out on the edge, or even whether it is driving business to the cloud, but the bottom line remains.

“It’s all about maximum uptime and this calls for an all-inclusive approach to ensuring there are more than one data center, cloud or service provider on the ready and fully capable of taking over the operations should anything happen.”

NonStop taking outages? I don’t think so! NonStop users needing to consider ways to accelerate recovery? Again, no sir! It’s all in hand. And with this I just have to add one final note – I am fully charged-up on this topic and it’s one we all need to champion whenever we are given the opportunity to speak up. And make sure we conclude every conversation with NonStop, a resounding No! to outages, planned or unplanned. 

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...