Your cloud journey – where should you start?

If you are reading this expecting me to tell you the answer then I am afraid I will disappoint you. Every single organization I have spoken to is different and I expect your organization is too. The cloud journey can take many different directions and what is the right place to start for one organization might not be right for another.

Beware of anyone that tells you their view of the “right answer” without first understanding your company, its priorities, its history and without first showing you the breadth of options available from which you can make your first steps.

This is a journey that needs to be navigated in partnership. It is your cloud journey not mine!

Early advice

1. Make sure you understand the breadth of what you can do today in the cloud.
2. Make sure you understand there are many different ways to work with the cloud. When you think about your cloud journey remember it is not a one size fits all.
3. Think about what your first, second and third workloads might be to decide who will be the best vendor for you.
4. Try to avoid falling into the trap of ending up with 10 cloud providers. Remember that learning multiple clouds is not in your best interest longer term and you will probably consolidate later.
5. Do something now. Your competitors are.

This is the third post in this series.

• The first one posed the fundamental question as to if you are falling behind on cloud?
• The second post tried to share some quantifiable information on what was happening in cloud land from the likes of Gartner, McKinsey, KPMG and IDC asking if adoption was really happening.

Today we will look at the outcomes of some of that great research and summarize how companies ARE moving to the cloud. In doing that you will get to see what options exist for your cloud journey and where others are putting their focus.

Despite what many people think moving to the cloud does not have to mean uprooting everything you are doing from day one. Also – the move to the cloud is a shift in operating model rather than a move to a “new place” and so it needs to be considered in that context.

The choice of location?

One of the biggest things to contend with in your organizations IT future is the decision as to if you want to:

1. Predominantly run on premise.
2. Mix-up on premise and the cloud (Hybrid Cloud – which is more complex than this simple description).
3. Move to public cloud (sovereign being a special version of that for certain countries).

Your chosen direction depends on many factors and I personally think we will see a mix of 2 and 3 as the de-facto IT landscape soon!

A quick simple example

To try to show you a bit of the thought process I have got a simple example abstracted from  a real world situation:

1. Hardware needs to be refreshed under an existing custom application. Two main drivers:
   • The hardware is old and has needed constant ongoing maintenance which costs a lot to maintain the warranty. The Hardware vendor is getting ready to pull support and pushing me to buy new tin.
   • The custom application is now running slow given there are a greater number of users and more data than when it initially went live. It is known that we will see a dramatic increase in users shortly due to an acquisition.
2. It is incredibly expensive to procure the new hardware required to scale to meet our new needs and the IT budget would take a major hit just when IT is being asked to cut budgets
3. Most of the original developers of the application have left so the focus is on just getting it to a new platform rather than re-architecting at this point
4. There is no alternative off the shelf application in the cloud that can replace this custom application.

In this case:

• Point 1 above says you have two choices. Buy new hardware,  look to get the application infrastructure from another source or look for a commercial off the shelf (COTS) replacement. Keeping going with what is there is not an option as the risk of failure is climbing. Yes – you could negotiate with the hardware vendor but you are really fighting a longer term losing battle.
• Point 2 above rules out the idea of going and buying new much better spec’d hardware which can scale to meet the anticipated need in the future.  This points strongly to the idea of getting the application or its infrastructure  to be provided by someone else or looking for a COTS replacement.
• Point 3 above rules out the idea of splitting up the application such that some of it van be run on premise and some in the cloud. This kills the Hybrid idea. It also rules out the idea of replacing parts of the application with more modern services. This leaves us with a complete move to an infrastructure provider or looking for a COTS replacement.
• Point 4, above, rules out a COTS replacement meaning the last option is to go find someone who can provide that infrastructure .

If we get to that point then we are talking about using something known as IaaS (Infrastructure-as-a-Service) for that application so you are consuming infrastructure not building and maintaining it. Your choice of where to get that infrastructure service from in the cloud remains open however.

This is on of the reasons why Gartner says:  By 2020, a Corporate “No-Cloud” Policy Will Be as Rare as a “No-Internet” Policy Is Today in their report on cloud you can see extracts of here.

A methodical look – A more complex, but similar, scenario

If we expand the simpler example I provided then we can look at one of the basic questions asked in the IT-as-a-Service Cloud Survey run by Mckinsey. They summarized the results in this nice graphic I have linked and included below. Their survey question allowed for the status quo to remain as well as virtualization strategies and private cloud thus expanding the choices and the complexity.

 

There are many conclusions that can be drawn from this single part of the survey. McKinsey does an excellent job in their paper that presents these results. I am not going to try compete with that here as you can go read it yourself. It is clear for all to see that everything is increasing in percentage terms on the consume side while it is clearly going down on the build side.

Conclusions on where!

One conclusion therefore is the “where” question is being answered comprehensively as “The Cloud” and therefore by definition “On Premise” is slowly being phased out. It is probably fair to assume many organizations are looking to get out of the infrastructure game and are looking at a way that is considered acceptable to them.

The same McKinsey paper breaks down the size of the company to see if that has any impact. It becomes clear that the shift is most dramatic in Large Enterprises who you might think are most risk averse. I think it is also as those are the ones running up huge bills for non-value adding infrastructure and hence they are moving the quickest!

Later in this post I will share a little more of what people are moving there.

So now you know IaaS – Say hello to PaaS and SaaS?

Today almost everyone knows that IaaS (Infrastructure-as-a-service),  where you are more or less managing everything from the virtualization layer up,  is one of three choices that you have when thinking about loud. So far in his post we focused a lot on that. The two other options as you go on your cloud journey are.

• PaaS (Platform-as-a-Service) where you are generally being provided a set of services you can integrate to build an application which you can configure and manage yourself (simplified description).
• SaaS (Software-as-Service) where you get a finished application you can fire up and use. Maybe in some cases you can extend that application too. This often gets a lot of the news headlines and it is what many people think about when they think cloud.

I will resist the temptation to provide examples here to avoid naming products. While many existing applications might go down the IaaS route first, it is also increasingly clear that new organizational needs are being assessed for suitability in the PaaS or SaaS world for a variety of reasons.

So what are companies moving to IaaS, PaaS and SaaS?

In their report “Leaders and Laggards in Enterprise Cloud Infrastructure Adoption” McKinsey shared fascinating chart. The chart looked at the main place different cloud workloads will land over the next few years. It provides some insight into the cloud journey other organizations are on from which you can draw some inspiration.

Here is a basic summary (You can get directly to the survey image depicting what is summarized here by following this link):

• SaaS: Back-office support, communication and collaboration and marketing and sales. This perhaps reflects where the current crop of mature SaaS applications exist.
• Public IaaS/PaaS: Development and testing, product development and management and the web. This also makes sense. It suggests a lot of movement is to reduce the costs associated with having to have permanent dev/test environments to maintain and scale. It also suggests that hosting web infrastructure today is far better done on a globally secure, and infinitely scalable, cloud than most can ever dream to achieve in house.
• Private IaaS/Paas: Core operations and Infrastructure. Perhaps reflecting that while moving to the cloud is vital there are some workloads people want in a totally segregated environment.
• Traditional: The largest workloads will be Core Operations with the rest all dwindling.

If you look at the IDC view of the world they hold a similar opinion. Here is an excerpt from this report that you can read in full here:

“SaaS, which accounted for 70% of 2014 cloud services spending, will continue to dominate public IT cloud services spending, as most customer demand is, understandably, at the application level. PaaS and cloud storage services will be the fastest-growing IT cloud services categories, driven by major upticks in developer cloud services adoption and big data–driven solutions. IaaS will remain the second largest IT cloud services category — boosted largely by cloud storage’s 31% CAGR from 2014 to 2018 — even as intense price competition and looming consolidation pressure IaaS price points.”

So what can you conclude from this?

1. Almost everything you are doing as a company is on its way to the cloud. If you are not moving your competitors are. They will soon have an advantage if they do not already.
2. Your cloud journey can start in ANY of these places.  The choices are endless and driven by your organization. Bottom line is you have to make some choices now! If you have got to that conclusion then this post has served its purpose.
3. Not all cloud vendors are equal. When you look at this list you can clearly see that some vendors are only able to help you in one of these spaces only. The implication is they push you there first when it might not be the right place to start. Meanwhile others can support you on whatever choices you make today and provide a path to the future. This is important as it lets you reuse skills and knowledge over time shortening and lowering the risk of future projects.
4. In the public cloud IaaS and SaaS are dominating today but PaaS is on its way to the big time!
5. The traditional approach is almost dead on its feet. Will you be the last organization breathing life into it and having to claw back savings others are making elsewhere?

Unscientific Twitter Poll

To conclude I did a simple twitter based poll on December 6th. In that poll I asked people to share their organizations plans around cloud for 2017. As I wrap this post up the results stand as shown below. You can see for yourself that in this small sample 45% aim to increase their usage of the cloud in 2017 and only 11% intend to do nothing in the cloud in 2017.

For me it is no longer about if you will move to the cloud but rather when.  The question then becomes if you will move quick enough to survive! To those already on the cloud journey… see you out there!

Up Next

In the next post I will expand on this mass move to the cloud. I will cover the motivations organizations have going in as well as some of the unexpected, positive, outcomes that some saw when they got there on top of what they were originally looking for.

As always I welcome your feedback.

_{Although I work for Microsoft these thoughts are 100% my own & may not reflect the views of my employer.}