Colin Horne | NetApp UK Ltd | IP Expo
Colin Horne Head of UK & Ireland Cloud Service Providers at NetApp UK Ltd joined us at IP Expo to talk about clouds and all the different capability and different functions that come with each one. Hosted by Bill Mew and Savannah O’Hare
Colin: Yep, Colin Horne. I manage the service provider and system integrator business for NetApp in the UK and Ireland.
Bill: Okay. So as our first guest here from NetApp. Colin has come to talk to us a little bit about some of the extra dimensions that you need to think about in the cloud dimension. Many people are talking non-stop and we had interviewed earlier today about how you take applications to the cloud and how you make the cloud a central part of your business. One of the things that people possibly do not think about is their data in the cloud and where is best to put the data and possibly the cost or some of the hidden costs and many people are caught up by ingress charges, egress charges and all of the hidden costs that may be in the cloud. You guys have been thinking heavily about this and helping clients in this way. What are the things that clients need to think about?
Colin: It is really interesting Bill because when people talk about the cloud they do exactly that, they say the cloud but actually there are lots of clouds and when there are lots of clouds that means there are a lot of choices all bringing different capability and different functions. One thing is still the case is every year the amount of data that gets produced the single year, generally speaking, is as much as all the information ever created since the year before and when you think about exponential growth data has got to be taken seriously and all information has different values. So where historically there would have been a conversation with IT Managers with salespeople it would have been ultimately about delivering infrastructure against workload. The world has changed and is changing in such a rapid way. Now the conversations are more leaning towards let us talk about the service levels we need and how we manage that at the right price not even on an annual or five yearly basis it is monthly. So, when you talk in that way the flexibility of where your information is on a monthly basis is absolutely paramount. The reason why organisations are really looking at data fabric strategy to enable that.
Bill: Not thinking about your data is not only something that people have always needed to do but possibly being a little remiss at because quite often they have been hoarders of data and having really differentiated on what data needs to be on immediate demand what needs to be archived or whatever. But some data discipline has been brought about by the arrival of GDPR and are thinking about exactly what you are doing with data. What are you doing with your data fabric strategy to help people actually address the sort of challenges you get with GDPR and actually having a better handle on what type of data you have on what you are doing with it?
Colin: Bill, this is generally put under the banner of policy. So, these are for organisations which need to set a policy on how they manage their information, GDPR is just one of many. You look at different industries and they all have different regulations. Of course, GDPR has got the big press right now everybody is talking about it. We all know about it because we are getting emails, right? So, we are dealing with it. Nobody even knew what GDPR stood for so and what was next. Then something that is regulation is set by policy. The most fundamental important factor here is you have got to have the flexibility on what you can do in your information and your data at any given time and that is essential which again, I am sorry. I will get back to that same term again organisations need to think about having a data fabric strategy, it is all about that flexibility of being able to move information around seamlessly between clouds on premise with service providers. Maybe in a few year’s time, we will be talking about cloud brokers rather than so sliders, maybe even data breakers. So, it is an interesting time that we are moving into. Once that change has happened, that will be a secular change, it will not turn back again. That will be the way in which we access our information.
Bill: And of course, there is no single cloud that is the best fit for all workloads and types of data, we are living in a multi-cloud world here and actually cloud is not necessarily a given. There are some companies that have gone to the cloud and decided that certain workloads need to be brought back down and are looking at hybrid strategies. Are you looking at solutions that span the whole breadth of different clouds and the on premise and hybrid environments? And if so, what are some challenges of managing all of that?
Colin: There is a term that has been used more and more, certainly with the organisations I see on a daily basis and the term is multi-cloud. Whilst that might not gel itself there, any individual hyper scaler by definition, are not necessarily going to be talking about multiple clouds, but if you talk to organisations who are the consumers of information, they are going to want multiple clouds. Multi cloud environments are the key thing then if you have got your information and your workloads in the right place at the right time now multi-cloud could mean on-premise, it could mean specific capability of niche service providers and it could mean taking the value and benefit from various hyper scalars. One thing is for certain you need a mechanism to manage it and that mechanism in the next 12 to 18 months and beyond will be absolutely key to making sure you are getting the right information at the right place at the right price point.
Bill: Of course, at this moment in time, we have containers and other different technologies, which help you host applications on potentially a number of different environments or cloud destinations but we certainly have not got true workload portability yet. That is kind of still out of reach on the applications but up on the data side what are the real challenges in terms of moving your data around?
Colin: You have already raised a couple of them, there are going to be commercials cost inherent in doing that, not everybody will favour seeing information move away from their environment, but I think we will see that just pure economics and commercials alone will drive organisations to adapt to the right policy to enable companies to want to choose that because of their service levels and their price point, that will happen. Let us take it a little further. Let us take it maybe five years from now it might not even be that far out. Imagine a world where you do not even need to consider thinking about where you are going to put your data. Imagine a world that is completely orchestrated. Automatically orchestrated as an umbrella that sits over all of your information that is constantly going to the public defining exactly what SLA you want and getting the right price point automatically put in there, so you can sit back knowing that every single month your information is in the right place, meeting your SLA at the right price point then you really do know that you have hit a panacea.
Bill: Sometimes we are talking about an environment where it may not be just your own data that you need to query there may be third party data set and people have termed things like data gravity in terms of you needing to have a level of proximity or questions about the latency around key data sector, an example might be the genomics England database which is the single largest genomic database. If you want to do some genomic Big Data analysis, you are probably going to want to locate somewhere near those data sets, so it can be a data-centric environment. Are we going to move away from that? Are we going to be completely agnostic or is latency in data gravity going to have some point to play?
Colin: There will always be scenarios and we could probably think of more of those where there will be a need to have information sitting as close as possible to the point of compute. For latency reasons that is just one example of hundreds of different ones. Now, where that is the case is there will always be, I say always probably always, be a need and desire to operate in a world that does not necessarily let us just go into the cloud. Okay, where they exist then again, you still may need the requirements, the flexibility to have access to information outside of a localised-on premise environment and moving information back in again. Again, we move back down to that same data fabric policy because that if you have a base fabric strategy you have that flexibility to do that and that is going to be really really important. For example, in the environment that you gave information has a life cycle we knew that many years ago. So, whilst you talk about important information in the example you gave how long is that information important for, can information become less important over time. If it does it needs to sit that close to the compute if not then why not take an economic viability of putting that into a cloud and you know it is there is going to be have no set answer to any set situation. Every organisation is going to have different information requirements and it is having the flexibility to be able to do that which is going to be key to these organisations going forward.
Bill: Looking further ahead. What are the big challenges on the horizon that you think we should be looking out for now?
Colin: Well interestingly and I think, this is my opinion. I do not think anybody wants to own anything anymore. That is the interesting thing. So we move away from a mentality of owning assets having liabilities on the balance sheet and depreciating them. So actually what is it we need, we need to have access to information, regularly, at the right price point by SLA. So, if you are moving to a world an entire world of not owning anything anymore, lets use an analogy like a water company. You would not go to your water company say give me a year’s worth of water. You might get that wrong. You might actually oversubscribe, how much water you want might cost you less money. What happens if you want more water and you have not got it, then you do not have water for three months while you sort of PO out and get it shipped to you. It is just crazy. It sounds like the most ridiculous thing you would do. So, why do that with information? So, become a commercial recurring cost that you have to access information in the right place at the right time.
Bill: I think if we run out of water, it’s not going to be a PO we will be worried about its probably going to be BO. Going beyond that what a NetApp up to at the show here today. I mean there’s an enormous almost number of stands you making announcements you doing any big presentation?
Colin: I leave that with other parts of our business to work on. I am just here as a delegate and meeting a few people and networking. So generally speaking our presence is pretty global. Most people know of Net App and hopefully, me sharing the love in todays session right here right now will help us a little.