Not so long ago, I was asked to give a 30-minute presentation to a varied audience on ‘How the adoption of cloud was affecting network solutions, and what the future holds’. I was happy to accept the mission but on thinking about the content, I thought it may be wise to take a brief look at the past and present, before getting out the proverbial crystal ball. A bit of research followed and I tried to keep the slides to a minimum and be as interactive as possible, always the best approach in my book.
“So, when do we think we saw the first great cloud event?” I asked the audience. I thought this would be a great way to break the ice and wake them up should the previous 300 slide presentation had the same effect as counting sheep. “2010” came the first reply, “Birth of the Internet” was predictably in the nominations and then came “Release of the smartphone in 1994”. This last suggestion sparked a couple of disagreements in the crowd, and I was pleased that the icebreaker challenge was most definitely in the bag. Presenter 1, Audience 0.
So, with the game afoot, I announced my opinion on when we saw the very first major cloud ‘happening’ and declared that it was way back in 1660. I was actually quite relieved that nobody had said anything earlier than 1660 or my whole chronological plan would have been in jeopardy!
It was at this point I saw many of the participants look up from their phones while multi-tasking, which is always a good boost for presenter confidence. I decided a long pause was appropriate at this time (similar to how reality TV shows make you wait for the winning contestant result) and then stated that “it was the General Post Office of course”. From various website information, I was aware that very well documented descriptions existed of Mail services going back as far as ancient Egypt, but I had to draw a line somewhere or this presentation would have been in danger of seriously overrunning.
In 1660 the General Post Office in England was established and available to the public, and I motioned to the audience that in my opinion, this was truly the first cloud revolution. Imagine the ability to post a letter with a short delivery timescale anywhere in the country for what was undeniably a relatively small fee. The alternative up to that point was to get on your horse (if you had one), and deliver it in person or alternatively pay for the dedicated carriage which was comparatively extortionate in terms of costs.
This public service was a revelation for family and friends, but for business? Well efficiency just went through the roof. For the techies in the audience, I considered an analogy that is often used when teaching the OSI layer. If the letters could be likened to IP Packets, then the standard Mail service would be similar to the UDP protocol, and TCP would be more similar to sending a Telegram (proof of delivery of said letter). The nods of approval from the audience and the zero loss of people since I walked on gave me confidence that the event organiser was not going to pull the stage trap door lever.
I started on my next chapter, “Fast forward 200 years or so to 1870 and the next cloud jump came in the form of the Telephone Exchange”. Imagine businesses across the country being able to discuss trade in ‘real-time’, not to mention what it meant for the emergency services. Telephone customers were now described as ‘Subscribers’ and even up until the 1990’s shared ‘party-lines’ were still in operation as the original build of infrastructure was far short of delivering one dedicated line per subscriber.
The rapid deployment of Telephone Exchanges was a major boost for employment, especially with each call being manually patched through by rows of ‘switch-room’ staff. Unfortunately, this brief surge in switch-room employment was short lived with the Automatic Telephone Exchange being developed in 1891, I guess you can blame Mr Strowger for that one.
In the 1960’s we saw 2 major clouds appear almost at the same time. Of course, the Internet was one of these, very much in its infancy, but the dominating business cloud was created by something called a mainframe computer. Large organisations, particularly Banks and financial institutions adopted IBM’s new super-computer architecture, and we saw a significant ‘catch-up event’ take place in the world of data communication. As the distance between terminals and Front-End Processors grew, the demand for data connections became a priority and the telephone companies used the existing copper-based infrastructure to create ‘private line’ transmission technology.
I finished this with the argument that this was one of the first instances of Cloud development influencing the demand for new network technologies. I could almost see the cogs whirring in the minds of the of the audience and I guess many knew what was about to pop up next. The element of surprise had now passed but, hey, it was good while it lasted.
The Personal Computer came along only 20 years or so later, and with it came the ability to host applications, generally one server for each individual application. IT Managers went crazy and servers were popping up at almost every company site, under desks, broom cupboards and basements. This created a huge dilemma for the Telecommunications Network with the exponential number of point-to-point private lines needed, and with this demand, newer any-to-any technologies emerged to create a more efficient handling of the traffic. ATM Frame-Relay and then of course MPLS were developed and this brought with it a new profitable ‘build one and sell to many’ platform for Telecommunication companies.
The audience was now one step ahead, and an eager front row participant (my favourite as soon as he raised his hand) stood up and declared that the difference between these events was halving each time. Bravo to that dude, even though he had stolen my thunder.
Server distribution was abolished circa 2005 (I made that date up but it seems about right) and the buzzword was now Data Center consolidation. Ethernet in the WAN was the next follow up technology to provide layer-2 LAN type services geographic locations, even intercontinental in some instances. The Metro Ethernet Forum emerged as a new standard and companies binged on global layer-2 connectivity for a multitude of services and catering for upper layer protocols other than IP. The pace of change was really heating up now, but the biggest change was yet to come.
At this point in time and my presentation, businesses had been used to a single dominant cloud being in place, even more so when the integration of voice over data was not an option any longer. Telecommunications infrastructure had been growing and developing, with each upgrade providing greater core intelligence and industry M&A activity creating huge operational assets.
The changes in the preceding few years showed a consistent gradient in terms of change, but nothing revolutionary, however the Clouds were looming.
I decided it was time for another one of those pauses for the audience, and then start with something that sounded like a line from the Terminator II film (ok a bit dramatic but it was my stage after all)
I wasn’t sure who made the first strike or on what day, but before we knew it Software as a Service, Infrastructure as a Service and Single Sign On, were upon us. The sheer movement of applications to the public or private cloud was no longer hype but considered a default position unless regulation intervened. Virtual ‘x’ was taking a strong hold and CIO’s were literally walking the Data Center corridors looking at what hardware could be virtualised and offloaded. Opex based models were of course a big friend of the CFO and the rapid increase of Global Cloud Service Providers meant that any Enterprise expansion was covered from a scalability and total cost of ownership perspective.
Almost overnight, the single cloud that businesses had been used to for so many years had dispersed into multiple clouds and it caught out the major Telcoms players, particularly in the Global Enterprise arena.
This overwhelming transition coupled with a dramatic rise in mobility meant that traditional MPLS networks were just not able to provide the freedom and flexibility that was now necessary. Hybrid MPLS/Internet designs were advancing to the max but still had major challenges and it was only a matter of time before the Cloud incited its next opportunity to rise. We now know that to be SD-WAN. I guessed that the audience could be extremely wary of any in-depth SDWAN discussion, as that topic has far from any shortage of content and hype in the market, so I was going to stay at high level as best I could on that subject.
This latest change of multi-cloud, hybrid public and private architecture has brought about a huge disruption to the market, probably long overdue. In this new world of Digital Transformation, CIO’s and particularly CDO’s are now faced with a plethora new suppliers and technology choices in the network arena. Global Enterprises, who had traditionally consulted Global Telco’s on WAN services, have now shifted to speaking first with new influencers in the form of SD-WAN vendors and Cloud Service Providers.
Systems Integrators have realised their potential to succeed based on their confidence to adopt software and virtualised network services and have broken into new areas of managed service demarcation points. In addition, we can see traditional towers in IT strategies being broken, the vertical responsibilities being challenged by new cross functional capabilities in SI’s, Telco’s and OTT players.
Telcos have faced yet another challenge in the form of financial modelling. Their intelligent core ‘super assets’ are under threat from the new, largely edge based intelligence that SD-WAN brings, making the network a commodity purchase for some customer profiles. This is evident in the surge of global network multi-sourcing for public and private regional connectivity suppliers.
At this stage in the presentation, I had poetic licence to declare that it was a great time to be a Sales Engineer or a Consultant. I was of course heavily biased, but positive disruption must be good yes?
Due to my ramblings I was near the 30 minute mark now and in terms of predicting the future, I had to admit my crystal ball was having some essential maintenance done. However, as we had seen, the massive rate of change that we had seen during the presentation suggested that the latest developments will not have time to settle, they won’t get chance.
The advancement of 5G, IOT and application-based networks will drive changes, up to the point where you can’t break the laws of physics. To add to the diversity, Public and Private based applications will always need a differentiation in service levels for guaranteed quality of service, with regulation being stronger than ever influencing Private Cloud IAAS and geographical dependencies. Nods of approval suggested I wasn’t on my own with the closing statement.
They say that out of chaos comes order. During this period of continued Cloud adoption, Global Enterprises need to have independence, stability and dependability, particularly when network multi-sourcing is deployed, and applications are migrating to SaaS and IaaS models. In lieu of this, it’s no wonder that a new breed of specialist providers has emerged to give Global Enterprises the confidence to fulfil their digital cloud transformation, and have recently been classified as Software Defined Interconnect Provider (SDIP’s). InterCloud’s Global SDIP platform has been developed specifically to simplify demanding and complex Cloud environments, its no match for a working Crystal Ball, but its the next best thing 😊