You're currently viewing Softcat.ie, would you like to continue?

Yes, I want to view Softcat.ie
No, take me to Softcat.com
Explain IT: Season 2, Episode 8 - Unstructured Data

Playing now - Unstructured Data

0:00
29:24

How do you manage unstructured data? 

In this episode, we dive into the world of unstructured data. We look at exactly what it is, the best way to store it, and the challenges we face when it comes to security and recovery. Andy Hardy, EMEA region vice president of cloud storage company, Nasuni, and Dylan Foster-Edwards, Softcat’s head of the office of the chief technologists join host Michael Bird to look at what the future holds for unstructured data.

From L to R: Andy Hardy, Michael Bird, Dylan Foster-Edwards
Host:
Michael Bird
Michael Bird Digital Marketing Manager Softcat
Guests:
Dylan Foster-Edwards
Dylan Foster-Edwards Head of the Office of the CTO Softcat
Andy Hardy
Andy Hardy EMEA Region Vice President Nasuni
Key takeaways
  • Unstructured data is information that users can edit, use and save. Users can decide on file names, how they organise them and how they access them. This is different to structured data, where the software decides where the information goes.
  • The amount of unstructured data we store is growing exponentially, accounting for 80% of enterprise file data. To relieve pressure on physical storage space, there are three options for managing your unstructured data.
  • The first option is a fully on-premise, dedicated solution, the second is a hybrid approach of on-premise and cloud, and the third is a 100% cloud-based solution.
  • These options are suitable for different areas of industry, and there are different benefits for each. It’s important to look at what might work best for you and your industry and objectives.
  • Moving unstructured data into the cloud opens up the ability to collaborate, worldwide. Future cloud-based file systems will vastly reduce cost and risk, whilst improving user efficiency.
  • The need to recover data will never go away, but cloud-based storage gives you a rapid way of accessing previous versions, without having to store secondary backups.
Solutions

Andy Hardy: Everyone's experience in their personal lives appear trivially easy to move data around the world, but frequently these days when people come to work we’re back in that siloed, clunky IT environment and I think to myself, “gee whizz, why isn’t this just easy, like in the cloud?”

Michael Bird: Hello and welcome to Explain IT, brought you by Softcat - the show for IT professionals, by IT professionals that aims to simplify the complex and often overcomplicated bits of Enterprise IT without compromising on detail. I'm host Michael Bird and over the next 30 or so minutes I'll be challenging our panel of experts to take a different area of the IT ecosystem and of course explain it. In this episode we’re going to be taking a look at unstructured data and with me to help explain, explore and understand is Dylan Foster-Edwards who is Softcat’s head of the office of the chief technologists. Now Dylan this is your second time on series two of the show. You brought an interesting fact before, what is your interesting fact?

Dylan Foster-Edwards: My second interesting fact is, as a part-time job I am actually an officer in the RAF. I ran a squadron for about 12 years and recently with work and children and stuff I’m now part of a training team who runs a couple of exercises twice a year.

Michael Bird: Nice.

Dylan Foster-Edwards: Keeps me busy in part-time.

Michael Bird: And with us to help we’ve also got Andy Hardy who is EMEA region vice president of Nasuni. Andy what is your interesting fact?

Andy Hardy: I have a small aeroplane, a Piper Cherokee, which I love. But I've flown that little Piper Cherokee all the way from here to Sydney Australia, my hometown.

Michael Bird: My goodness, how long do that take?

Andy Hardy: Seven weeks and two days.

Michael Bird: Wow. Did you have any hairy moments?

Andy Hardy: Crossing the Timor Sea, which is like five and a half hours of water, we were halfway across that and the engine just coughed and stopped.

Michael Bird: Oh my goodness!

Andy Hardy: It stopped for about eight heartbeats and then it started again.

Michael Bird: Wow that must have been nerve-racking! What would you have done?

Andy Hardy: We had passed an oil platform about 80 miles previous and I remember selecting that in the GPS, cos I'd marked it as we went past, you just do. And I turned towards it, I realised I could glide about 20 miles and I was thinking, could I swim 60? And by then the engine had sorted itself out, we had some bad fuel that had an issue.

Michael Bird: Andy, also what does EMEA region vice president actually do?

Andy Hardy: I run the sales organisations. We have people in the UK, in Germany and the Netherlands, in Ireland and we have sales, pre-sales, technical support, professional services that type of thing, so I’m responsible for the P&L here in Europe and East Africa.

Michael Bird: So Andy, first things first - unstructured data, what does it mean?

Andy Hardy: Unstructured data - most people just call it files, but we use the word unstructured because users are able to put stuff and organise stuff where they want it and how they want it. So structured data would be like in an application or a database and the software is deciding where the information goes; unstructured data is when users or groups of users can organise themselves around that data.

Michael Bird: How does that differ to other types of data?

Andy Hardy: So structured data would be like block data, SAN storage, the simple data storage that's needed behind databases and applications. File data is what we put in NAS -network attached storage - things like NetApp and Isilon and those type of products. So the fundamental difference is that users get to name the files, decide where the files get put, decide how to organise a hierarchical directory structure so that you can work as humans like to, which is fuzzy, it's not always rigid like a database table. So it's being able to have not just your personal files stored somewhere, that's a fairly trivial problem that most people would be familiar with from their laptop or cloud storage for their personal files, but in an enterprise environment it's really a focus around group shares where lots of people can access files.

Dylan Foster-Edwards: And I guess it's an area that I know we see from customers, it continually grows because they never delete anything, users never tidy anything up, so it just grows and grows and grows.

Andy Hardy: Any IT director will tell you that no one knows what to delete, or is afraid to delete - staff leave, “can we get rid of that?” “I don't know, let's keep it just in case,” so unstructured file data is the fastest growing and the largest portion of enterprise file data, it’s about 80% of all the data by terabytes that's in people's environments.

Michael Bird: What are some of the challenges that organisations are facing with unstructured data?

Andy Hardy: The fundamental issue is this interminable growth. In the old days we used to be able to stick all the data on one disk, or one tape reel, or something like that and then that wasn't big enough, so the disk got bigger, but they weren’t big enough so then we started stuffing the disks in a box which you could call a NAS, like a network attached storage so I could get more data in one box, but the fundamental problem still remains that one day that box will fill up, so I’ll buy another, what we call, shelf to stick in the box and then one day I won't have any room to put more shelves and that’ll be too big so I'll throw all of that away, literally stick it in a skip and I'll buy another one, a bigger one, the next generation one, the one with more compression, or whatever, and I’ll stuff that into my basement, which some people call their datacentre. This cycle has gone on for several decades now and we've reached a point for many businesses where just stuffing more data in a box in the basement or in my ROBO sites, my remote offices and branch offices, we’re just running out of the ability to do that. People don't know what to delete, what can we get rid of? It's a challenge.

Michael Bird: Can you give some examples of some different industries?

Andy Hardy: I mean if you think about manufacturing, people have really critical information, CAD - Computer Aided Design drawings - 3D modelling, things like in the construction industry things like Revit models that tell us how a building’s put together and what all the different components of that building are. You need to keep that for a long time, obviously because of building maintenance. In the media space you would have, if you're producing a magazine ad, the images, the text, the graphics for that. If you're producing a TV commercial you would have all the raw footage of that etc. All of this is unstructured data. If you're in the oil and gas industry, a lot of our customers are in oil and gas, then you’re collecting humongous amounts of data in the form of seismic returns that need to be analysed, in the form of drone images of rigs at sea and undersea camera, video of pipeline inspections, there’s helicopters flying around this country all the time inspecting power infrastructure and the video from all that is kept because in the old days we would collect that raw data, analyse it and just keep those analysed results, but increasingly what people have understood is that, for example with seismic data, we might need to revisit that raw data at a later stage with greater insight than we had initially and reanalyse it. And so there's this great reluctance to throw it away, it's valuable and we might need to work with it again and again. But having said that, the vast bulk of what we store will never be accessed again, we don't know which needle in the haystack we’re going to need later, but that's the nature of the problem across a range of industries, it just grows and grows and it's business-critical, valuable information and if I can't get it, it may be that I can't do my job, that I can't satisfy a customer process of some sort and of course that affects revenue.

Michael Bird: Why is that data so hard to manage?

Andy Hardy: Unstructured data is a growth problem, it’s a capacity problem, and we haven't yet talked about really how you store it, but historically it's been in your own premises and so at some point you're going to have to expand those premises, build bigger data centres, invest in technologies that somehow compress it, and alongside that I've got to do something else as well besides merely storing it. What if I lose it? How am I going to get it back? And there's a few different aspects of that backup and restore process and business continuity and disaster recovery that we have to take into account. Long story short, we need to make another copy of all that data, in fact we make several more copies of that data, what’s called secondary data, so that we can recover a past version of a file. Another challenge associated with that backup process is the backup window can take more time than we have to make that secondary copy of data. So we've all sorts of techniques for trying to speed that up and snapshot approaches and so forth, but fundamentally, people run out of time to make that backup copy. And then lastly if I'm making a whole copy of my files and computer systems elsewhere for disaster recovery in a DR site, then I've got the problem of how am I going to fail over to that as well as the cost of having all of that secondary data. So these challenges all multiply as we grow this, and it's growing exponentially.

Dylan Foster-Edwards: I was going to say about growing is an issue because the technology we've got is increasing the size of the files. We’ve been working a bit with police forces, and they have body cameras and they’re just moving to the next generation of body cameras, and the data increases about four times the size that was previously.

Andy Hardy: And another problem that goes with all of this data is what we call retention, so it may well be that I might need to look back at data from a very long time ago and it's really hard for me when I'm storing this stuff as an IT guy to know which data and when. But if you think of the problems around medical processes, clinical imaging for example, it may well be that someone's condition recurs and we need to go back and get past images and compare them to now and of course in a clinical setting I kind of want that past image really quickly, and that's quite difficult to do. And then there's a fresh challenge, and that’s around security and people getting their data corrupted by bad actors - people who wants to ransom money from you because they want to take control of the access to your data by encrypting it, potentially, and other challenges like that.

Michael Bird: So for an organisation then, what options do they have to manage this unstructured data?

Andy Hardy: There's three main options. First of all I could have it all on premise in standalone NAS, so that's network attached storage, that's storage that has a file system built into it. Secondly I could have a hybrid approach where the solution is partly on premise and partly in the cloud and then there's also a third approach which is to have a 100% pure cloud solution.

Michael Bird: Ok so let's talk through then the option of fully on-premise.

Andy Hardy: Sure so if I'm an organisation with a main office with 1000 users let's say and I might have a dozen or even hundreds of smaller offices with smaller numbers of users, what I would do for the on-premise solution is I would put a large NAS array into the main office because I would have a lot of people with their data in one system. I’d probably have a replica copy of that either on the same site or perhaps at a remote site or a separate disaster recovery site to protect me from that risk of losing that site, or losing that system. I would also have to implement at that large site back up and then I would also have to have similar solutions, different sizes, perhaps smaller in my ROBO sites, my remote offices and branch offices, those systems will let users access files in that local system quite readily but it typically is the case that to access the files from far away from another site I would use the WAN, and that's a very expensive thing, a wide area network, and it can also be very slow, as the files get larger it certainly gets slower.

Michael Bird: Ok so that's fully on-premise, and then what about the hybrid model?

Andy Hardy: The hybrid model would entail having part of the solution on premise, that would typically be the file storage, and then they might have cloud elements for the backup and it might be that in some of their ROBO sites, the remote offices and branch offices, they decide that they can do without having a local copy of the data and access those files from the cloud. If it's just personal files that people need then it's trivially easy to get it in the cloud, but if it's a large amount of data for those sites then that might be a more challenging problem, but there are ways to stitch together as people have started to figure out how to get the infrastructure into the cloud, a combination of on-premise and cloud.

Michael Bird: And then finally all cloud?

Andy Hardy: Well all cloud is something that's growing in popularity and when I say cloud I might be talking about public cloud - vendors like some of our partners like AWS and Microsoft Azure and IBM Cloud Object Storage and Google GCP, Google Cloud Platform, but equally when we talk about cloud it may be on-premise, so there are cloud storage solutions that you can run in your own facilities and share globally as well that come from a number of vendors, people like Cloudian, in for example, so the idea of a purely cloud file storage solution has some great advantages. First of all if all of my files are in the cloud then I know that all the data is protected. If I have all my files in the cloud then I can share it more readily around the world.

Michael Bird: Can you give some strengths and weaknesses and maybe some reasons why an organisation would pick one or more of those different options that you just talked through?

Andy Hardy: I think you would stay with an on-premise solution, or perhaps even a private cloud solution where you just weren't comfortable. And this might be the case with some investment banks for example, that your file data, your commercial data, your customers’ data is going to be in someone else's infrastructure. That will be a good reason to keep the data on premise. You might also choose a hybrid solution because your approach to solving the performance problem is to keep the hardware local and then have the backup and the secondary data copies in the cloud. The problem with that of course is, with both of those aproaches is you're still left with the fundamental issue that you will run out of capacity and you'll have to buy more and you have to invest large amounts of capital in order to grow and expand and duplicate that data. And then lastly the reason that you would put data into the cloud would be primarily to reduce the cost, to increase the accessibility of that data from any part of the world and to be able to enable staff to collaborate globally rather than just from within one office. Increasingly some CIOs are all taking the view that perhaps these large global cloud organisations can do maybe a better job of protecting that data as well than they are able to do, or are able to afford to invest to do on premise in their own infrastructure, so the old idea was, is it secure in the cloud? The new idea is how can I be as effective at keeping it secure on premise as some of these large global cloud vendors might be able to do? But then we've got some other trade-offs to consider.

Michael Bird: So what are the disadvantages of going all cloud?

Andy Hardy: So in terms of data sovereignty, it's very often the case, sometimes for regulatory reasons, sometimes just for purely corporate politics issues, that my German operations might decide that they really want files to be stored in Germany. Maybe that's how their customers feel about that data. And it might also be that I've got operations in China where behind the great firewall of China, as some people call it, the Chinese Government will sometimes insist that your data is held in China. What I'm going to struggle with, in those circumstances, is putting all of my data in the same place. So one of the things that's important for customers is that they are able to put data in more than one cloud environment at the back-end. Now what we don't want to do with that is then complicate how users access that data, there are technologies like Nasuni which can help you with that problem which can present it as a single global file system and yet in reality we might have it stored in different parts of the world for those data sovereignty reasons.

Michael Bird: So is what you're mentioning, is that basically an enterprise file-sync-and-share system?

Andy Hardy: No, so a lot of users will be familiar with enterprise file-sync-and-share systems, OneDrive for business will be extremely well-known, things like Google Drive and Dropbox etc and technologies like ours which is a cloud based file system are complementary to that. So what file-sync-and-share does, is it makes sure that all of my files are accessible, not only on my laptop, but on my iPhone, in my home office and all of that. That is a way to put unstructured file data in the cloud. But if I think of the larger problem and it's most of the terabytes, most of the petabytes that enterprises have, it’s the files in the group shares in the NAS, and put simply, I can't get rid of my NetApp filers my Isilons etc, by copying all that data onto your laptop and having you share it a bit with your mates. One, it won't fit. Two, that puts it on you to share this stuff around with everyone, so it's really a bit of a bifurcation in the market at the moment, and so we see that personal data goes typically for a home drive onto your laptop and your file-sync-and-share, your OneDrive as part of the Office 365 Suite etc, and on the other hand if I have this cloud-first objective that many CIOs have, the NAS is like a boat anchor keeping me in the datacentre, and what stops me putting it all in the cloud typically is that it's going to make it slow, it’s going to be far from the users. So you can use techniques like ours where the file system moves to the cloud and we combine that with caching of files on premise to do that transparently and in a way that I couldn't do with file-sync-and-share.

Dylan Foster-Edwards: So I guess one of the challenges that we see from some of our customers when we start talking to them about, well have you considered moving some of your data into the cloud? Especially in some sectors like legal sectors as an example, they get concerned around sovereignty, they get concerned around privacy. Why don’t you talk just a bit about how what you guys do can help or support or alleviate some of the challenges that they may feel by moving that data that they’ve got their arms around in their datacentre to someone else's datacentre, or in the cloud?

Andy Hardy: You’ve touched on something really important there, which is data privacy. And I think, for a lot of customers, it’s more important than even the data security consideration. Here in Europe, the big concern which affects all Europeans corporations and lots of North American and other global corporations that have operations here is GDPR. And it requires that if we have someone like a third party access that data, that we disclose that so the concern that immediately springs to a lot of people's minds is, well if I'm putting all this data in the cloud, how am I going to keep that data private? And there's a couple of different ways that people look at that. You’ll hear lots of people talk about, it's important that we keep data encrypted in transit and at rest. So keeping data secure in-transit is using things like TLS encryption over the wire. Keeping data encrypted at rest is just a question of encrypting it and then storing it. The part that people I think really should be concerned about is where has the data been encrypted and who has the encryption keys? And a lot of encryption at rest schemes, the default ones in the cloud, mean that the encryption keys are actually in the cloud environment along with the data. Now if we just take a step back, historically what led us to GDPR was concerns in Europe about the USA Patriot Act. The USA Patriot Act says that the cloud vendor, they're typically American corporations, is duty-bound if the US feds, or other agencies ask them to disclose the data, to give it to them and to give it to them decrypted. So they will subpoena the cloud vendor, they will get a copy of the data just by asking and that cloud vendor must not tell you that they've done that. Of course both of those things - not disclosing it and doing it in the first place - are in contravention of our regulations here in Europe.

Dylan Foster-Edwards: To be honest, most of the law firms have been the reason why they’ve avoided it, so they're looking at solutions to ensure that no one can read the data without them knowing about it.

Andy Hardy: The silver bullet here is pre-encryption of the data before it goes to the cloud. And there are technologies, including ours, that automate that process and make it transparent to users, so the data is encrypted with the customer’s own encryption keys that the cloud vendor has no sight of.

Michael Bird: So for the end-users then, how do all these different options affect them and what is it that you think an end user is looking for when we're talking about unstructured data?

Andy Hardy: On the whole, end-users kind of don't care about the infrastructure, they just want it to be invisible to them, but I think there are some aspects of this that impact end-users. First of all, everyone's experience in their personal lives of using cloud services make it appear, seem and it is, for the user, trivially easy to move data around the world. But frequently these days when people come to work it doesn't really work like that. We're back in that siloed, clunky IT environment and I think to myself, gee whizz, why isn't this just easy, like in the cloud? So users’ expectations have risen, they just want to be straight-forward and easy and that's one of the reasons that increasingly CIOs have a cloud first strategy on top of all those commercial and technical advantages, we want a user experience which is cloud-like. The other thing that users care about when they're at work is their ability to collaborate with other users. Think about a building design model, a Revit model or a 3D CAD model for a design of a car component or something. I might have some of the right engineers in London and I might have some of the other engineers in Kuwait and others in India and I might want them to be able to collaborate on all of the same things at the same time and of course, moving unstructured data into the cloud opens up the ability to do that and some vendors like us have technologies for global file locking that enhance that and make that that way of working seamless, globally.

Dylan Foster-Edwards: I guess one of the things that we see - and manufacturing was one, we work with a couple of car manufacturers and construction companies. They’re always complaining that, as you say, the users just don't really care, they just open the file, they save it and they're not worried and the end up having multiple copies of the same thing all the time. They would love to get to a position where there can just be one copy and it can be shared wherever they are whether they’re opening it here, opening it in another country. So there’s never a duplication.

Andy Hardy: So that is one of the key advantages of doing what we've done at Nasuni, which is to move a file system from the edge into the cloud itself. One of the things that lets us do is just have that one gold copy of the file and all of the deltas for that file, no matter where in the world that they came from. We can keep that as an infinite number of changes to the file and that lets us do a couple of things. One is we solve that commercial problem of not duplicating data. Two is we can enable users to collaborate globally and thirdly it means that we can go back to past points in time quite rapidly.

Michael Bird: So is what you're mentioning, is that basically an enterprise file-sync-and-share system?

Andy Hardy: No, so a lot of users will be familiar with enterprise file-sync-and-share systems, OneDrive for business will be extremely well-known, things like Google Drive and Dropbox etc and technologies like ours which is a cloud based file system are complementary to that. So what file-sync-and-share does, is it makes sure that all of my files are accessible, not only on my laptop, but on my iPhone, in my home office and all of that. That is a way to put unstructured file data in the cloud. But if I think of the larger problem and it's most of the terabytes, most of the petabytes that enterprises have, it’s the files in the group shares in the NAS, and put simply, I can't get rid of my NetApp filers my Isilons etc, by copying all that data onto your laptop and having you share it a bit with your mates. One, it won't fit. Two, that puts it on you to share this stuff around with everyone, so it's really a bit of a bifurcation in the market at the moment, and so we see that personal data goes typically for a home drive onto your laptop and your file-sync-and-share, your OneDrive as part of the Office 365 Suite etc, and on the other hand if I have this cloud-first objective that many CIOs have, the NAS is like a boat anchor keeping me in the datacentre, and what stops me putting it all in the cloud typically is that it's going to make it slow, it’s going to be far from the users. So you can use techniques like ours where the file system moves to the cloud and we combine that with caching of files on premise to do that transparently and in a way that I couldn't do with file-sync-and-share.

Michael Bird: Let's talk a bit about the future then. What do we see as the future for unstructured data? Is it all going to be in the cloud?

Andy Hardy: Look I think that CIOs are driven by reducing cost and improving user efficiency as well as reducing risk and I think that we are heading rapidly towards a point where I'll be able to minimise risk, certainly minimise cost and have the best user experience without having any on-premise infrastructure, I think that's really where we’re headed. Today I still need some on-premise infrastructure but the more I can reduce that, the more I can shrink that down, the less capital I have to spend on that, the less risk I take in managing that. Whether that cloud destination is on my premises in one location using an on-premise cloud technology stack or whether it's in Amazon's basement or Microsoft’s basement, is kind of a moot point, and so for me, the future is that the on-premise infrastructure data centre evaporates, it goes away.

Michael Bird: Then if it's all in the cloud, does that mean that because the platforms are so reliable and so secure that we won't then need to back any of that stuff up?

Andy Hardy: People will always need a way to go back to how data was at a previous point in time. The question is what is the most efficient way to do that? One of the most pressing reasons to do that is ransomware. We see that it's becoming a growing threat so in the worst-case my ultimate line of defence is being able to go back to how that data was before that encryption event took place, so that's just one reason why the need to recover data will never go away. However I think that the need to make a secondary copy of data again and again and again, which is the way we’ve historically done backup, I think that will go away because what we’re able to do, particularly if you have the file system in the cloud, the file system is able to track all the deltas for all of the files, then I have a way that I can very rapidly, in fact practically instantly, go back to a previous version of a file or a whole bunch of files in a project, or a whole share, a whole volume, all of the volumes if I need to, to that past point in time. We're able to do that however without making that secondary copy and spending all that money. The technique has existed for a long time, it’s snapshots, it's just tracking deltas, but the constraint historically, has been that I could only track those deltas to the limits of the box or the disk that I could put that in in my NetApp array or whatever in the basement of my office. So the big change that I see is the ability to provide really rapid business continuity even if I've had my files encrypted or we just bug it up and change them inadvertently or whatever it may be, even if I lost a datacentre because the techniques that the cloud vendors use mean they're resilient to the loss of servers, racks, even whole data centres if you pay enough, that obviates, for me, the need for traditional secondary storage, so I say don't do secondary storage for file data, it's done and it's expensive and it's slow.

Michael Bird: Ok so from a user perspective, let's talk about the future then. Will there be a scenario where users will basically just hit save, the computer intelligently knows where the user would want it to be saved, you know, the folder whatever?

Andy Hardy: Well for personal files that sort-of happens now with technologies like OneDrive and that's great. But I think that for group share, the power of unstructured data is that we get to choose as a group or perhaps as a business process, how we're going to organise a process, how we going to organise our data. So for example I'm not sure that I foresee that AI would decide who’s going to change the plumbing in my 3D building model and who's going to colour in the trees. I need the freedom to say my enterprise is going to this a little differently to the enterprise down the road and I don't think that that commercial advantage is something that we hand over to AI. I think AI, of course, can speed up processes, can give us fresh insights and so forth but for me the power of unstructured data is that it's unstructured. That really means that users are deciding how it's structured and applying a structure from their own minds to that data instead of having a machine do it.

Michael Bird: Ok so to summarise, Dylan can you find a bit of a summary?

Dylan Foster-Edwards: Yeah sure. So we've covered a lot of topics with Andy today so I guess we started off with a definition of unstructured and structured data and the differences between them. We talked about the types of data, whether it's personal data or group shared data and why you store those in different places. We talked about different types of technology that are available, so your more traditional on-prem, NAS-based storage, which most organisations use these days, moving to your more common enterprise file sync and share technologies that you get with the likes of OneDrive, Citrix Share File, those sort of things that people are undoubtedly using. And now we’re looking at the benefits that could be achieved by moving to more cloud file services, and the benefits of that type of technology can give our users. And we talked around removing that challenge of users and having multiple copies of the same file and how, by using cloud-based technology, we can simplify that and give those guys ability to reduce the complexity and reduce the need to hold multiple copies without actually the user needing to understand how it works, they just save it as they always have done before and it deals with it in the background. We talked around how we can help with the GDPR challenge; obviously we’ve passed that deadline now but people still need to be compliant and if you're using the analytics and search capability that this type of technology can give you, would enable you to find that data that you need, especially as, if you're going to store that in the cloud, you’ll most probably have a  larger set of data, so it's going to be even harder to search it than it would on-prem. We talked about the management of your unstructured data and how that could reduce the cost and complexity together with increasing the flexibility to access your data, we talked about having cached versions of your data in one location or whether it's all in the cloud, you can pick and choose which one you want, really, ultimately, you can have the technology to meet the need whatever you need in each of your locations and also it is about trying to fix the problem of your group shared data, manage that in a long term strategy.

Michael Bird: Dylan and Andy, it's been really interesting talking to you both. Thank you very much for your time, thanks for coming in today. Listeners, if there's anything in this show that has been of interest, or if you'd like to speak to someone at Softcat about anything that we’ve talked about in this episode, do feel free to get in touch and we’ll include some links in the show notes. Please also do make sure you click subscribe wherever you get your podcasts and we’ll deliver the next episode to your device as soon as it lands. So thank you for listening to Explain IT from Softcat.