Blue Chip – LinkedInLIVE – March2020
Blue Chip – LinkedInLIVE – March2020
Chris Smith [00:00:08] Hello and welcome to today’s special episode brought to you by Blue Chip. I’m Chris Smith from Blue Chip. And today, we’re going to talk about IBM’s latest storage announcements. I’m joined by Lee McEvoy, storage expert from IBM.
Lee McEvoy [00:00:22] Thank you, Chris.
Chris Smith [00:00:23] Hello, Lee. So IBM have recently started to use a tag line, simply smart. So what I wanted to do is explore how simply smart when it comes down to these announcements from IBM. So, Lee, could you expand upon what these announcements mean for the viewers?
Lee McEvoy [00:00:42] Absolutely, So like a number of storage vendors out there, IBM’s had a portfolio of solutions and in common with it, a number of our competitors, often we’d have more than one solution that could be fitting a customer’s requirements. And that level of complexity in terms of when you’re positioning to a customer can have uncomfortable consequences for the customer that they might: why am I picking this versus the other? And also, I witnessed this in a previous role at a company I used to work for. It can lead to in fights between the salespeople at that vendor who might only get paid if it goes one way versus another. And I was there, a customer environment that’s needed refreshing that we helped design a solution for. And absolutely, IBM, having multiple options, did feel slightly uncomfortable for the customer. They just wanted IBM to say this is it, which absolutely did help them understand why we’d picked the one option that was the right fit for them. But they were very unimpressed by the other vendor who happened to have a big fight in front of them about which of the two options was the right one for them. Simplicity also has some additional advantages. If you are developing one product family effectively, you put all your resources into it and you make it the best it can be. And it also allows you to focus your resource and minimise the cost of the resource. So you actually able to provide a better solution and a lower cost to the customer. Simples good.
Chris Smith [00:02:16] Simple is smart. Apparently we’ll explore that in a little bit more detail going forward. So for me, this announcement is quite significant. The most significant one for a while in terms of how they changed it. So we touched then on how it’s simplified, the infrastructure that some of the previous announcements that I’ve been involved in have seen before came not just around the infrastructure, the underlying technology, but came around the software side of things. So the functionality that comes along with the hardware as well. So I think back to when the Spectrum Suite was initially released and the flexibility that delivered a kind of decoupling a lot of the functionality away from technology and delivering it straight on to whatever technology IBM had at the time and allowing you to software define the storage and effectively that’s enabled us to build a software defined data centre at Blue Chip. We’ll cover that in a bit more detail later. But how does the simplification of the hardware match up with the simplification around the spectrum suite?
Lee McEvoy [00:03:29] OK. So each of the storage hardware platforms had its own storage software component as well. And as you’ve touched on very clearly there, IBM’s had a software defined view of the world for a long, long time. And most of our, if not all of our products were available to be bought as an appliance for customers that wanted that level of simplicity and ease of use or for customers that wanted to customise it or potentially run it in a slightly different way or reuse some of the assets they already had. There was the capability of buying the software and effectively building your own custom environment. And that’s actually been a very positive thing for us because that investment that we made has allowed us to move to enable customers to work in a hybrid multi-cloud way as well. So good example, a customer that used one of our spectrum virtualised based solutions, things that we used to call store wise. But now parts of the flash system, family and effectively a customer can buy one of those boxes, will have one of the historic ones and they can replicate to an instance that’s running an IBM cloud or in A.W.S also running spectrum virtualise. So software defined allows customers to run the solution that they need to run the most effective way and gives them flexibility, especially as they’re moving to public cloud.
Chris Smith [00:04:49] Little bit of independence comes along with it as well.
Lee McEvoy [00:04:52] Absolutely. And I think that software defined the way of doing things means that customers can effectively commoditize the hardware layer, which means that they tend to pay less for their storage and often. Customers will ask me, why is only B.M. commoditizing the hardware layer, because that makes it less likely that you’ll buy IBM storage. Actually, in our experience, it makes customers more likely to buy IBM storage because some of the innovations we make in that space still save customers a huge amount of money. But it changes the vendor lock-in that has been historically a large part of the storage world where customers once they bought into a platform, they have to buy the upgrades from that vendor and that vendor alone and in that price negotiation from procurement. Where’s the power? entirely with the vendor being software defined actually allows the customer to have that level of control pulled back towards them. And as I can tell from personal experience, Blue Chip use that very, very well when negotiating for new purchases from IBM storage.
Chris Smith [00:05:58] Yes. Thank you very much, IBM. So from a technology point of view, then when I when I look at where IBM was in the past and one of the things, you know, I was a massive fan of at the time was when we bought store-wise Products, while the functionality might be somewhere down the development path. When we originally bought the systems we had, when that functionality became available, we could utilise it even back, you know, several years ago when when we were buying these systems, that kind of thing was just starting to become available from IBM. And now I think it’s come to maturity. I would say with this announcement. But back then as well, I think also there was there were still a number of products across all of the IBM technology stacks across whatever you were using. There were multiple different options you could take. Now, I can see IBM strategically removing themselves away from having all these multiple options, which basically all the main competitors have, can set them apart from that. Because let’s be honest, I think when it comes down to storage. IBM has lost a little bit of its market share over the last few years. I think it’s fair to say, but I think this is a great opportunity strategically to change that because they really are setting themselves apart from the rest of the market space in what they’re doing. So do you feel that this is this is going to deliver what you need? Is IBM to change that?
Lee McEvoy [00:07:28] Absolutely. I think if we look at the primary storage market, having the ability to go to a customer and say we’ve got the right solution for you, 6:09 valuable hybrid cloud ready, flexible enough to do workloads, whether it’s bare metal, virtualised containerised going forwards. And obviously with IBM’s investment around the purchase of RedHat, we’ll be leading around open shift and that ability to run workloads seamlessly on public cloud various and to have the interoperability to move it without becoming locked into a public cloud vendor A versus B versus C, but also to decide to bring certain workloads back at the right time or burst out that investment. IBM is made around poaching. RedHat is absolutely key and will bring IBM the kind of benefits that the owner of VM Ware used to experience as well. But if we look at the flash system announcements and saying I can now go to a customer and say I can give you that level of availability, I can give you a very, very simple, easy to use storage system with all those capabilities that you’re after. And really all we need to talk about is how much capacity you want, what level of performance you want, and then we’ll come up with the right size engine running that spectrum virtualised platform, and you’ll get to get the benefits of IBM’s investments around the hardware being differentiated as well, potentially around flash call modules. It’s a very simple conversation. We’re not talking about do we want it, this, that or the other. We’ve got one platform that is a good fit for all primary workloads and it’s very easy for us to explain. And that makes our marketing life a lot easier, which means our messaging will be a lot more effective.
Chris Smith [00:09:17] Okay. So I think back to the point earlier, is simple smart. I think the simplification of of of the whole range of storage is the thing that jumped out at me anyway. So I think it is a smart move. I think it’s clearly market leading as well, which changes the game a little bit out there, puts the IBM storage ahead of the competition as far as I’m concerned. I’m slightly bias because we obviously own quite a bit of IBM storage. So we wanted to be successful. We want this level of innovation as a business ourselves. So talking about that and talking about Blue Chip where we are. You’ve been heavily involved. Thank you. Over many, many years now. Our growth from a storage point of view. So IBM, sorry Blue Chip as an IBM platinum. Business partner, we have expert level competencies in storage as well. So can when I look at what we’ve got on the ground at the moment, is it something like 100 plus same controllers, over 200 drawers of desks, 14 s the C’s, that sort of thing? There’s a huge state of storage within Blue Chip. You know, you because you’ve been involved, you have some technical insight into that. So. So can you explain give some examples of how we’re utilising that?
Lee McEvoy [00:10:37] Certainly. So, you know, it was slightly embarrassing when we found out how the enclosures that I helped you acquire over the years because it was quite a scarily large number. But if I give some examples over the past few years, so approximately three years ago, you had about 2 petabytes of storage across your two data centres. Now it’s closer to 11 petabytes. Most customers experienced double digit growth, but double digit starting with a 2, 3 or 4. It’s really exceptional to have people growing double digit, but closer to starting with a 9. Yeah. So that’s 500 percent. Yeah, 500 percent over three years is very, very significant growth. And what I can tell you is you’ve not used up anything like that amount of extra rack space in your two data centres. So effectively what we’ve done is we’ve leveraged the new technologies that IBM have delivered. And if we look at the primary storage, the most recent acquisition that Blue Chip made around primary storage was to use some of IBM is now flash system products and to effectively be able to put the IBM flash call modules in what was a V 7000 new product equivalent would be a 70 200 flash system and put that into what is effectively just two year Rackspace and to effectively be able to get millions of eye ops and huge amounts of bandwidth and incredibly low latency out of that. And I think the last the last box that was put in had approximately three hundred and seventy terabytes of usable space before data reduction. And the amazing thing about IBM is unique classical modules is the compression capability that they have. It’s always going to do it. Why do we always do it? Because we don’t induce any further latency. We don’t limit the performance so effectively. The latest generation boxes that can have almost three quarters of a petabyte usable space with data reduction that could easily be three 4 petabytes into year of Rackspace. So that’s how we’ve managed to contain the massive terabyte growth and can keep that in a small number of rack units. What you’ve also been able to do is take advantage of IBM’s acquisition of Clever Safe. And you’ve got IBM Cloud Object Storage, so that marks you out as being fairly specialised. Not many organizations, not many service providers have that yet. In the arena that we’ve been working in and effectively that makes you a true cloud provider, because if people have got cloud native applications, they want to be able to write object because that’s how you do unstructured data. So using that IBM cloud object storage in your environment for that coupled with another use case which is around the backups and having an active archive area at a very attractive cost per terabyte for your backup environments, for your customers when you’re doing backup is a service or as part of a wider offering really, really key. And again, it’s driving up the efficiency because of the clever aspects of IBM cloud object storage. And it’s also doing it in a very, very dense rack unit space and very power efficient as well.
Chris Smith [00:14:00] Yeah, and I can see how that investment is just not enabled us to deliver more services to our customer and it being more flexible. But actually the commercial aspect of it as well, which will cover in a bit more detail in a moment, has enabled us to do it at the right price point for our customers. What customers expect. Ultimately, when you’re operating in a cloud environment, people will be comparing you to some of the larger cloud operators from a cost point of view who have huge scale. I mean, we think our scale is large, but these guys have a much larger scale, but we’re able to compete because of the way IBM operated. And the I guess, the innovation within their storage portfolio. So in terms of what we’ve invested in, what we’ve built what a lot to think of as well, as I mentioned earlier, previously we could future-proof ourselves because we buy the technology, be ready for the next innovation that comes along. I think from what we’ve invested in, there’s some of that bill in as well, isn’t there?
Lee McEvoy [00:15:02] Absolutely. And one thing that’s probably worth highlighting as well is. Customers requirements. And when you bring them on board to your cloud, often aren’t what the end point or the future requirements are. So for a really good example, a lot of customers will come on board and say, yeah, we want to have two sites just replicate the data. And then over time what they realise is actually what they really want to have is a H.A. system with automated fail over, because most people look at the two metrics, whether they call them the proper names of the two metrics RPO Recovery Point Objective and RTA Recovery Time objective must be want to get to zero and pretty much zero. And Blue Chip have made an investment primarily around the networking because our storage doesn’t actually charge more to do the hate chain solution, but the investment you’ve made around the networking is enabled your customers to migrate from. Actually, I’ll just take the replicated solution to being H.A. and effectively aiming towards an RTA with near zero. And that’s something that Blue Chip have chosen to do. And because you’re in a software defined data centre world, you’re able to make that migration for customers very, very simply. Tell me about the other elements about future proofing. There’s two things that are front of mind for customers when they’re buying storage at the moment. And that’s in M.V.E and actually M.V.E and then V.me over fabric’s, although they’re not particularly well understood. And also storage class memory and effectively M.V.E and we may over fabric’s in storage class memory are all about saving money. And that sounds a bit strange because most people consider it to be expensive. And actually compared to traditional flash storage class memory is expensive, but the amount of money it has the potential to save you for your workloads that are expensive to run because of software licences, potentially workloads such as well known databases that can save money storage is cheap compared to licencing expensive databases. So that’s an area that we’re very keen to work on. M.V.M.E, you’ve got M.V.ME. Storage, you’ve got a lot of M.V.M.E Storage. And in fact the flash systems that Blue Chip were purchasing for the past four or five years, they weren’t M.V.M.E inside, but there was something that was just as fast. So you always had the ability to go sub millisecond. And in fact microsecond latency for key workloads and you have customers that took significant advantage of that. The next piece of future proofing that you’ve already covered is the ability to go in M.V.M.E over fabric’s. And at one really clear example of where Blue Chip were future proofing and Blue Chip with the first company in the UK, not just service provider, the first company fullstop to for a 32 gig capable sound director environment in the whole of the UK which the guys at Brocade now Broadcom were very proud of. But what’s M.V.M.E over fabric’s? And why is it important? Most people think it’s about just making the storage faster. Sometimes it does make it a little bit faster, but fundamentally what it does is it saves the customer c_p_u_ cycles on their host where they’re paying for virtualisation software, operating system, software or application or database software. And if we can say to a customer, I can give you 20 percent of your performance back just by making this change and then you get to do 25 percent more work load on that without increasing your software cost or buying any faster CPUs. That’s quite attractive for a lot of people. And Blue Chip are ready for your customers when they’re asking for that.
Chris Smith [00:18:58] Yeah, and I think Blue Chip in general always look to reinvest back into our technology, into our datacenter consistently as you have experience to improve the solutions, improve the performance and improve the experience for our customers. So thank you for that as quite in-depth from a technology point of view. So the other bit that you touched on is, is Blue Chip software defined capabilities, and we mentioned that earlier when we spoke about Spectrum Suite and what that’s able to do. So, you know, what I can say is that that the storage we procured from IBM has allowed us to deliver software defined datacenter that can provision the not just the storage. But the compute the networking and the software as well. All all all at once, all very, very quickly or very, very flexibly. You know, we use all sorts of scripts and playbooks and orchestration to deliver that. But the IBM storage that the spectrum suite part is a key component of enabling. Enabling us to deliver that. So it’s really, really important to us. But I think the other part you mentioned as well, which which is is is quite close to our hearts. Blue Chip as well as is how we able to reduce the amount of space that our storage takes up. So the march in advance of flash and its capacity and its capabilities has actually enabled us to reduce some of our storage footprint over the last few years. You know, we’ve reclaimed not only Rackspace, but we’ve reduced our power consumption around that as well. For Blue Chip, that’s really, really important. So we run very ecologically friendly data centres. We use free air cooling. We don’t use harmful refrigerants or traditional computer room air conditioning. So we run peewee of 1.1 3 at the moment. So that means for every amp we use to power, the system is it’s 0.1 of an amp to, sorry 0.3 of an amp to cool it. And then, width course, as you know, offset power usage by planting lots and lots of trees in a forest. But I think, again, if we if we look at what’s what’s enabled us to do some of this, it’s the IBM storage. It’s the flash storage. It’s the density of that storage. Now, we were talking about some figures earlier. It was within a two year. You can get what now in terms of,.
Lee McEvoy [00:21:32] So, three quarters of a petabyte usable before data reduction. So with the compression that’s built into the flash call modules, that’s about a petabyte and a half. And if you want to use the other data reduction capabilities, such as deduplication as well and get further savings, so there you can go for petabytes, two year rackspace
Chris Smith [00:21:52] So in two years Rackspace, you can get four petabyte flash storage. So super fast, super efficient. I think in reality you can’t can’t get that much greener than that. So, you know, it’s made it a massive difference for us at Blue Chip.
Chris Smith [00:22:10] So so in terms of what we’ve done as a business, I mentioned earlier that we’ve invested heavily in IBM storage technology and we’ve future-proofed it and we’ve been thinking about the future. I think as well it’s not just about what we do with the technology. So we bought it. We run it. We own it. We have the experience of running it as well. But we also supply to it to end users as well. And you sometimes have been involved in doing that as well. So it’s not just about what we do. I think as an organisation, how we run it is our capabilities beyond that in advising and supplying storage to our customer base. But it’s not just a here’s a new system. Good luck with it. We can deliver it, configure it, help them run it the whole life of their storage, if you like. We can support them on because we were a user ourselves. We know what it’s like to use this technology and we can we can advise customers, which I think is a big differentiator for lots of businesses out there. Sometimes they just buy off of a traditional partner who will sell them their equipment and come back in three years and sell them some more. Whereas we’re there for the whole life, life, life term of that. And you’ve been involved in some of that historically, haven’t you?
Lee McEvoy [00:23:25] Absolutely. And I think the credibility from being able to say, well, I’m running effectively a cloud environment on the same stuff we’re proposing to you. And this is our experience and this is why we’ve recommended it. That does bring a lot of warm feelings to customers. But I think there’s also been novel ways that Blue Chip have taken IBM storage messages to customers not say subliminally, but in a quite interesting way. And the example I’d use is you do DR you have DR offering to customers. And as part of that offering, DR plan isn’t worth anything unless it’s being tested. So for the reasons you alluded to about saving Rackspace and power in your data centres, you actually moved your d.r test environment to be on flash, which was quite forward thinking.
Chris Smith [00:24:17] And Crazy as well from a cost point of view.
Lee McEvoy [00:24:20] But it was just it wasn’t as much as people thought it was going to be. But there was a traditionally DR done on the old rubbish that you’ve taken out of your primary environments just because it was free. But the power in the cooling isn’t in the rack space isn’t. But there was another angle that hadn’t been considered and there was a discussion with the customer that hadn’t been considered either. And that is there’s only so many weeks in the year and so many days in the week that you can do these DR tests over. So effectively having a customer come in and their system be up and running faster was a benefit to the customer because they got in, did their test and got out faster. And that meant you could do more customer tests in the time. We had the secondary benefit to that was a good proportion of the customers were like, what are you running on? is this some amazing system that you’ve got in the bag?
Chris Smith [00:25:18] Why is it so quick?
Lee McEvoy [00:25:19] Why is it so much faster than my environment? And they expect me to say, oh, we’re on the largest possible power box, which are very good, by the way, in case anybody from power is about to get unhappy with the new power boxes are awesome. But effectively, the DR box didn’t need to be a very large box because people would work standing up systems. They weren’t doing performance testing on it. Not traditional performance test. The performance they really cared about was how quickly could I get on test? It all worked. It was all there and then leave. And effectively, that was storage driven. So having flash is your environment for the storage to recover on to from your virtual tape libraries basically got a lot of customers saying, oh, what’s your environment? Well, it’s a small power, seven box. And in the first instance it was and it’s a flash system. And that led to an awful lot of flash system conversations with customers and some sales both for private cloud and sell through sales. And the ones I was involved in, I was very pleased with.
Chris Smith [00:26:19] I’m sure you are. I’m sure you are. So. So when we’re talking about customers procuring this wonderful newly announced IBM storage and making investment. So I’m thinking there are people that might have moved away in the past from IBM storage to one of the competitive brands. There are others available. However, if we look at this, this simply smart announcement and what this means to the storage industry, if you like, this is this is a big change. So if we’re trying to encourage people to invest in this storage and the new technologies, it’s not just about the features and function. It’s also about the cost. So what are IBM doing from a commercial point of view to encourage customers to make the leap away from wherever they are now on a competitive storage set back onto IBM or to use IBM for the first time?
Lee McEvoy [00:27:17] So what I’d say and I’ll cover this from a number of different angles if that’s okay. You say IBM has a reputation which is unfounded for being expensive. In my opinion. And actually, I’d say ask for a quote and compare it versus what you’ve got from somebody else. And you’ll almost certainly be surprised because, in my experience, our cost is lower. So that’s personal experience. Stop me and try and try me. The second thing I’d say is that actually most procurements now are done from a mindset of we need to be working in a cloud like way. And cloud is just somebody else’s computer potentially. But it’s also around the orchestration automation that you’ve touched on and being able to get things done quickly. But it’s also potentially about having an OpEx model for workloads that are variable. And so that’s where IBM have brought out something called storage utility offering, where effectively a customer can buy a percentage of a box and that can be the right size box for them and with the right amount of growth or headroom for them. And effectively, IBM then charges a cost per terabyte per month when they use above the initial purchase capacity and that’s charged quarterly in arrears. You know what? You’re going to get charged because what is being reported to you as well as to IBM and a number of other vendors have similar type models. But when I speak to customers about it, they’re always saying, well, is it like this one where I don’t get to? I’m committed to grow over time? Is it this one where I can only go one way some effectively buying the capacity as I use it? I can’t give any back what happens at the end of the term. Do I always have to lease it? Is it some kind of financial chicanery? And what’s really nice is to be able to say no, it’s not like that. No, it’s not like that. Variable variables variable is net, so variable up variable can go down. That’s why it’s called variable. Do I have to start small and grow big? No. There’s a lot of customers out there who have a strategy to move to public cloud, but they want an expensive platform now and they’re not ready to move. So if you need 100 terabytes now, but you think you’ll only need 30 in the future because of where the workloads going. What do you do now? Well, in a lot of cases, people are stuck paying very, very expensive, prohibitively expensive, ongoing maintenance charges for the hardware and the software from some of our competitors. Whereas what they could do is come to an IBM partner and say, can I have a hundred terabytes now stored utility. I only want to pay 30. Front and that variable will be going down as I move workload. Yeah, absolutely. We’re happy to go start large, go small. That storage utility is variable and at the end of the term, you own the asset. It’s not ours and it’s yours. Whether you’ve paid for any of those terabytes ever in this hundred terabytes scenario, you own a hundred terabytes of storage now.
Chris Smith [00:30:31] Hang on so. So if I buy a box and only use 30 terabytes of it is capable of 100 terabytes at the end of what period are we talking about here?
Lee McEvoy [00:30:40] Typically three years.
Chris Smith [00:30:41] So the end of three years, if I’ve only ever paid with 30 terabytes at the end of three years, all of a sudden I’ve got 100 terabytes.
Lee McEvoy [00:30:47] Absolutely. So, if you bought a hundred terabyte box, but you’ve paid for the first 30 terabytes of capacity, you have the right to use that 70 terabytes of burst capacity. And by the way, we don’t measure performance, but you’ve got the full performance of the hundreds as well. You get to variably pay a cost per terabyte per month for what you use above 30. And even if you never use anything above 30, you own the hundred at the end.
Chris Smith [00:31:14] Wow. Well, that’s a bit different, isn’t it? So. Yeah. And also quite simple as well.
Lee McEvoy [00:31:21] Well it’s quite simple. There is a degree of complexity. But this is positive for customers. And we live in a world where storage gets cheaper over time. So if my variable cost remain the same every year for the three years, that can be unpleasant for customers. So what we do is when we design this, we typically design it. So cost in year one is X cost in year two for the variable terabytes is X minus 10 percent and then in year three it’s X minus a further Y if you call that minus a further 10 percent. So I mean literally every time I’ve been asked a difficult question by the customer, I’ve smiled because I’ve got a good answer.
Chris Smith [00:31:58] So so okay. So I think let’s just agree at the end of this, that simple is smart. But what you’ve done is you’ve simplified the platform, you simplified the options around it from a software point of view as well. Made it easy to pick and choose what you need there. You’re also simplifying how easy is to buy it as well. And I think, stop me if I’m wrong, but in all three areas there, you’re market leading?
Lee McEvoy [00:32:25] Absolutely.
Chris Smith [00:32:26] There’s no one else doing this for storage at all out there at the moment.
Lee McEvoy [00:32:30] There are people that are doing elements of it, but not the whole big picture.
Chris Smith [00:32:34] You pull it all together. It’s a unique offering. So so I guess IBM’s put itself boldly because this isn’t without some sacrifices, I suppose. But they put themselves boldly upfront as probably the the the most leading supplier of of storage from a technology and a commercial point of view. All we need to do to try to customers to come and buy it. So I think from from from that point of view, you know, the way you’ve simplified it, the way you’re going to enable customers to reduce their costs by simplifying it and reduce the costs for management, reduce the cost of buy it in the first place. Based on what we were just talking about as well. If you couple that kind of opportunity in offering with what Blue Chip are able to do in conjunction with with IBM, which is to give give customers a degree of impartial advice as to what the best best direction to go in because we’re talking from a user’s point of view and enable them on that journey. Even if they do want to come to Blue Chip Cloud, we can supply them the hardware. So I think I think it’s great. I think it is market leading. Lee, thank you for joining us today for this. And thank you all for watching. I’ve been Chris Smith. He’s been Lee McEvoy. Thanks for joining us.