January 19, 2012 | Written by: Neil Weightman
Share this post:
Neil Weightman interviewed John Easton recently. The discussion concentrated on the future of IT departments, how organizations need to change to embrace cloud, the legacy of grid computing and how customers are currently engaging with cloud.
(Read the previous parts of this interview: Part 1 and Part 2.)
Neil: Could you describe the role of the Cloud Computing Lab?
John: That’s where I’ve just come up from this morning. The Cloud Computing Lab in the Galileo Centre at IBM Hursley is a place where we take a lot of our clients to show them our cloud technologies. It gives us the ability to showcase both our private cloud and some of our hosted cloud offerings to clients.
Neil: What kind of client were you speaking to today?
John: A large media client. They’re in an interesting position in that they want to implement a cloud in Asia-Pacific and our cloud platform will not be available there on the time scales that they ideally would need, so there are debates about whether they can actually slip their time scales to better meet our cloud or, more likely, that they’ll go and build a private cloud for this initiative and then move to an IBM hosted cloud at some time in the future.
Neil: Why did they particularly want it to be hosted in Asia-Pacific?
John: That’s where they’re rolling out. They have tens of thousands of users of a new platform that they need to deploy. They want the servers hosted in their local area. The speed of light isn’t fast enough – although CERN might be questioning that at the moment. The duration of time for data to transfer between the US or UK and Asia-Pacific is just too long, so you need a local presence for this. We have a cloud data center in Singapore, but it doesn’t support the cloud offering that they want at this point in time. There is a data center in Japan already and we’re also building one of the largest cloud data centers in China; this customer may be able to use the centers in Singapore or Japan at some point.
Neil: I understand that the Patriot Act can cause problems with certain customers if their data will be stored within the US. Are there similar problems with other countries?
John: The Patriot Act is often thrown up more as an excuse than anything else. Actually cloud doesn’t change anything: clients have been outsourcing their IT to us for many years and we’ve been subject to those rules in the past and we always seem to have been able to contractually get around them. Actually, it’s not a question of servers being physically hosted in the US, it’s basically the fact that IBM is a US company, so servers hosted by IBM anywhere are subject to it as with any other US cloud provider such as Microsoft, Amazon, or Google.
One of the interesting twists is that we have our cloud data center in Germany and the client we spoke to this morning said that Germany is for them not the best place in Europe to have a data center because of the data protection laws. They’re in the process of moving servers out of Germany because having them in the UK is easier.
Neil: Now that we have solutions like IBM SmartCloud Enterprise, is it easier to sell something like that to a business person as a package?
John: SmartCloud Enterprise is really an infrastructure as a service solution, which will give you a virtual machine running in an IBM cloud. You’ve still got to do something with it, because most organizations tend not to just buy virtual machines, although we have plenty that do. What business users are typically concerned about are the applications that they’re going to run and that’s where a lot of the SaaS [software as a service] capabilities are attractive, like LotusLive, and BlueWorks Live, which came out of the Lombardi acquisition. As part of the October announcement, we announced a suite of additional capabilities from the likes of CoreMetrics, Unica, Sterling Commerce – other acquisitions that we’ve made that have brought along their SaaS capabilities to enhance that portfolio.
That’s really the area that IBM wants to play because that’s where the greater value is over time – SaaS, BPaaS [business process as a service]. Our friends at Amazon have largely commoditized IaaS [infrastructure as a service] already and if they’re selling it at eight cents an hour or whatever, there’s not a lot of margin to be had in that area. Part of it is table stakes for playing in the cloud area – we have to be there, we have to have those offerings but in reality it’s the higher value things where IBM can differentiate itself – not just on cost but in actual capabilities. That’s where we really want to play with cloud.
Neil: Where does that leave Systems and Technology Group?
John: Interesting question. You could say “If I can buy my virtual machines from the cloud then why do I need to buy physical servers?” I’d turn that round and say “Well the compute has got to come from somewhere.” The volumes of data are going to be huge so you need storage as well to hold this stuff. There’s still going to be a requirement for servers and storage, and in that case, it might as well be our servers and storage than someone else’s. We’re actually quite successful in selling IBM’s hardware and software capabilities for organizations to build a cloud, out to service providers , telecom [telecommunication] companies and IBM Business Partners. Everyone seems to want to have a cloud offering, if they don’t have one already, and we seem to be reasonably successful in selling to those companies our core hardware and software assets, which has to be a good thing.
Neil: A couple of weeks ago I visited the cloud data center in Raleigh, North Carolina. It seemed to be mainly full of racks of servers. Where does the mainframe fit in?
John: For many organizations, cloud is synonymous with x86. Microsoft’s cloud runs on x86, Amazon’s cloud is all x86, Google’s cloud is x86, so that’s really why. IBM SmartCloud Enterprise is an x86-only cloud. IBM SmartCloud Enterprise+, which gives us a higher level of service, can run on x86 and Power systems running AIX. One of the key differentiators for us as both a cloud provider and a provider of technologies to help clients build clouds is that we’re not restricted to any particular hardware platform, so if a user wants to put a mainframe into their cloud, or AIX into their cloud and x86 into their cloud or different sorts of storage, our Tivoli cloud software stack will support that.
The Tivoli Cloud Management platform that we’re typically selling to clients allows them to mix and match the most appropriate technologies within a single cloud platform. Within a single cloud you could have an x86 platform that’s able to deploy VMware, KVM, or Xen virtual machines. We’’ve got clients that want to use Hyper-V virtual machines or LPARs in the Power systems running AIX space. They could be deploying VMs running either AIX or Linux. On the mainframe you could have Linux on System z. All of those could exist within the same cloud, all managed by the Tivoli software.
By having that flexibility, of being able to deploy on the right platform rather than being forced into one particular platform choice, then you have the ability to do interesting things regarding the cost of computing. We can deploy Linux on a mainframe significantly more cheaply than you can deploy it onto a lot of x86 kits, just in terms of the electrical power requirements – the mainframe can consume as little as 20% of the power for a given number of VMs compared to x86. The cost of energy is a significant factor. The challenge, particularly with large x86 server farms, is that the sprawl of physical servers consumes a huge amount of power and very often they’re poorly utilized, which is another way of coming back to the cloud virtualization discussion. By driving that virtualization hard, we can get away with fewer, bigger servers as part of the consolidation exercise.
Neil: Is it the SaaS offerings that are currently proving most interesting to customers?
John: We don’t break out how our cloud revenue comes in from the outside world.
Clients are still showing a significant interest in private cloud, where organizations want to build clouds on their own premises – that’s hardware, software, and services revenue – the traditional way that we do things, albeit deployed in a slightly different way.
There is a growing interest in things like IBM SmartCloud Enterprise and we seem to be doing quite well with clients, provided they understand what it is that they’re getting when they get into that space. IBM SmartCloud Enterprise+, our managed hosted cloud platform: there’s a huge amount of interest in that, but the challenge is that it’s not all there yet.
SaaS is an interesting one. A lot of customers in the SaaS space have come over from previous users of the software before we acquired those companies. As we bring out that portfolio and as we start to move the message higher up the stack the message becomes stronger. What I mean by that is this: We’ve been saying for a while now that cloud underpins a lot of IBM’s growth initiatives – things like business analytics, Smarter Planet, and really, until the October announcement, that was something we talked about, but the proof points were limited. Now we’ve announced the first four of those cloud solutions where you can directly see how cloud is starting to play into those Smarter Planet workloads, like Smarter Commerce, Smarter Cities, business analytics, and Smarter Social Collaboration. Those sorts of things make cloud much more real to clients who want to use it to transform their business.
About John Easton
John Easton is an IBM Distinguished Engineer and the Chief Technology Officer for IBM Systems and Technology Group in the UK and Ireland. He is internationally known for his work helping commercial clients exploit large scale distributed computational infrastructures, particularly those using new and emerging technologies.
John is currently the UK Technical Leader for Cloud Computing shaping IBM strategy in this key business area and helping clients with their implementation and adoption of cloud technologies.
He has worked with clients in a wide range of industries, predominantly banks and financial markets firms but he also has significant experience in the telecommunications and public sectors. Previous to his current role, John was the Chief Infrastructure Architect for a first-of-a-kind core banking infrastructure replacement program.
During his time at IBM before this, John has led IBM initiatives around hybrid systems, computational acceleration technologies, grid computing, and mission-critical systems. John is a member of the IBM Academy of Technology and a Senior Technologist in the IBM Innovation Network.