13 reasons that MIGHT prevent companies from successfully leveraging public cloud services

Cloud computing seems to be the holy grail, at least that is the impression I get when I scroll down my Linked-in and Twitter timelines. Don’t get me wrong, like everybody else I’m also excited about everything that is going on around ‘the cloud’ and the potential it brings to the table, never mind the type of public cloud service used. I mean, no upfront investments, flexibility, burst capacity (up and downscaling) at our fingertips, ease of management and maintenance and so on, what’s not to like? But, ‘the cloud’ is also just getting started, it isn’t ‘mature’ and/or seen as proven technology by many – there are still a bunch of reasons why companies might wait to leverage (public) cloud computing in general, go hybrid or perhaps won’t join the cloud movement at all, at least not for the first couple of years. I just wanted to put this out there and see what you guys think.

Disclaimer: the purpose of this article is not to dish ‘the cloud’, I’m not saying ‘the cloud’ is bad or that you shouldn’t make use of ‘the cloud’. ‘The cloud’ holds great potential and it will continue to evolve in rapid pace, I’m sure. I’ve come across numerous use-cases myself (in fact, I’m studying the Azure architecture as we speak, exam included) and I love the innovation going on between Citrix and Microsoft as well – now we’re getting somewhere. But… ‘the cloud’ is not for everybody, is has downsides as well, eventually not every company is going to move their datacenters into ‘the cloud’ and even in the case of hybrid deployments this will take time.

The general assumption

It seems to be a trend to think that within 3 to 5 years from now cloud computing will have been widely adopted by most organizations around the globe. And I’m not just talking about SaaS solutions like Office 365, no. Here I’m referring to on-premises datacenters being moved into ‘the cloud’, virtual desktops, applications and data included. It will be cheaper, less complex and as a result easier to manage, at least that seems to the general thought.

While this may be (partly) true in a lot of cases, it will depend on the type of (cloud) solution chosen and thus the problem you are trying to solve – which is also a valid point of discussion. Why do you want to move to ‘the cloud’, what are you trying to gain, how will you benefit, really? Considering ‘the cloud’ just because it will be cheaper, for example, doesn’t make much sense. Something not to be taken lightly, unfortunately this can be a hard message to get across.

I’m aware that there are thousands of companies out there who are already successfully leveraging cloud computing, no matter if it’s IaaS, PaaS and/or DaaS perhaps combined with a couple of SaaS based applications. Though the latter is by far the most popular cloud service on the market today. Also, a revenue growth of over 30% for Amazon including a stagering profit increase of 832% up from $92 million to $857 million of which most profit came from AWS, also tells us something. In fact AWS grew 58% year over year. The same applies to Microsoft Azure, the company states that its cloud business is over $10 billion, but that includes services like Office 365. Though Azure has grown over 120%, according to Microsoft and there is no reason to think that these number will decrease going forward, the exact opposite is more likely.

The cloud is taking over, but…

It is also a fact that for a lot of companies it is still (too) early. They have trust issues, data (integrity) issues, they worry about performance, internet connections (latency, bandwidth, all eggs in one basket etc.), about being in control, legacy (Windows) applications with tons of potential compatibility issues and more. At least this is what I’ve seen throughout the last year and a half or so. Even with one of the most popular cloud services today: Office 365 some of these ‘concerns’ arise, although throughout this article I am primarily referring to and focusing on datacenter and virtual desktop-like solutions hosted in, or from the cloud.

Ultimately I wouldn’t be surprised if it will take at least another 10+ years from now to get most of these companies into ‘the cloud’, if at all. It takes time. Virtualization took over 20 years to get where it is today and even now architects are discussing physical over virtual.

And while technologies like containers, Hyperconvergence, application layering etc. have been around for a couple of years now and everybody is talking about them, only a few percent of the worlds datacenters/ companies are actually using them. The same applies to cloud computing, it is still not seen as proven technology by many, at least not when it comes to large scale IaaS, PaaS like deployments.

Below you will find a couple of (potential) reasons why companies might choose to stay with their on-premises datacenters, at least for now or will only partly leverage public cloud resources going forward. Note that in the above sentence the emphasis is on ‘might’.

1. Company (-sensitive) data will always be an issue. Where is it stored, will encryption be applied (security), who has access (unauthorised access, privacy), who owns the data (responsibilities) etc. Depending on the branche/ type of company – this will continue to be a problem going forward.
2. Not all applications function well, or well enough, when physically separated from their data. Even with the latest developments around Azure Remote App/Citrix XenApp Express and the Windows 10 VDI from the cloud announcement, data locality will continue to be (very) important.
3. Internet connections. Your Internet connection(s) will become vital. For some it will feel like putting all of their eggs into one basket. Next to this, network variables like bandwidth, latency, jitter etc. will continue to play an important role in the overall achievability and end-user experience. Even more so when your applications and data are (physically) separated from each other, as mentioned previously. While solutions like express-route in the case of Microsoft Azure, or Direct Connect for AWS, for example might work wonders, they’re not cheap and it’s probably something you will have to purchase next to your existing internet connection(s).
4. While cloud datacenters (Azure, AWS) seem to be everywhere these days, customers are still dependant on how far they are physically located from one of these datacenters. Being (relatively) close will automatically mean faster response times etc.
5. A lot of companies are worried about user authentication taking place in ‘the cloud’. And even while this isn’t always the case technically (and when it is, it can be very secure) this can be a tough one to get across/explain.
6. Trust. Not only with regard to company data and authentication as mentioned above, but also when it comes to uptime, security breaches, SLAs, responsibilities etc. And what about performance and the overall user experience? A lot of companies still don’t consider the cloud to be ‘mature’, so to speak.
7. Exit strategy/ vendor lock-in. Companies are often worried that as soon as they have made the leap into ‘the cloud’ they can’t undo it. Or that it will be costly to do so.
8. A great deal of companies do not want to rely on a third-party handling their infrastructure, data and/or applications. They want to be able to have full control when they want or need it. Depending on the cloud service chosen this isn’t always possible.
9. Hosting your infrastructure/data centre in the cloud isn’t cheap (at least not today). Also, making the transitions from on-premises to the cloud takes time, testing and careful planning. It’s not something you do overnight. If ‘it will be cheaper’ is your number one reason to consider cloud computing, you probably need to re-think your strategy.
10. Cloud providers are not more secure by default. On average they invest more when it comes to security, but the bigger you are the more interesting you become. And eventually they will find a way. And if they do, the impact will be massive.
11. SaaS solutions are great, but they will only get you so far. We’re still dealing with tons of legacy (Windows) applications for many more years to come, which depending on their accompanying back-end systems and data placement can be hard to ‘cloudily’. In the end application and/or data compatibility and integration issues can make or break your cloud deployment. Startup companies, almost without exception will leverage SaaS services instead of investing in private datacenters, which makes perfect sense. However, not every ‘new’ company will be a startup. Many companies are based on the franchise concept, for example and more often then not they will simple have to use, software wise, what their HQ has to offer. The same applies to other horizontals like financial and insurance companies, healthcare, pharmacies, law firms, government institutions and the list goes on. These types of businesses all use very specific software tied to their field of expertise, which will take years to migrate, recode, cloudily etc. Eventually it will happen but this isn’t as straightforward as you might think. Just recently I spoke to a large government institution – almost all software related to their core-business is developed ‘in-house’, and they will continue to do so for many more years to come. Secure data placement is just one of the issues. This example alone impacts tens of thousands of users.
12. The same applies to DaaS. For a lot of companies DaaS will be a great fit. However, there will also be a bunch of companies for which DaaS will be too restrictive or limited with regard to what they can do or control themselves. Let’s also not forget that services like PaaS, DaaS, IaaS etc. are all slightly interpreted differently depending on who you talk to, this will include some of the (configuration and management) options you might or might not have to your disposal, which will differ per provider. Where are your user profiles stored, can they be customised, what about base-image control, user customisation, printing, scanning etc.
13. There has been talk about this for a while, but what if managing, supporting and setting up your on-premises virtual desktops/ datacenter is or becomes easier, cheaper and more flexible then its cloud counterpart? With all the advancements made in the Hyperconvergence space, and they are from done, implementing and maintaining your on-premises infrastructure is becoming easier and cheaper by the day. The days of weekend long firmware updates are long gone and ongoing management and maintenance is a breeze. Adding in additional capacity in the form of storage, compute etc. is as easy as adding in an extra node and the software (SDS) will take care of the rest. You, as a company will have full control and plenty of time to innovate and spend on other projects. If you look at Nutanix, for example they offer some unique cloud integration options making it a very valuable platform, since hybrid cloud computing will be the way forward, no doubt.

Partly sourced from Inside Citrix – The FlexCast Management Architecture

Other references used for this post: CIO.com

Bas van Kaam

Field CTO EMEA by day, author by night @ Nerdio

Father of three, EMEA Field CTO @ Nerdio, Author of the book Van de Basis tot aan Meester in de Cloud, Co-author of the book Project Byte-Sized and Yuthor of the book: Inside Citrix – The FlexCast Management Architecture, over 500 blog posts and multiple (ultimate) cheat sheets/e-books. Public speaker, sport enthusiast­­­­­­­­: above-average runner, 3 x burpee-mile finisher and a former semiprofessional snooker player. IT community participant and initiator of the AVD User group Community world wide.