Big Blue clouds are emerging on the horizon with the launch of IBM’s new set of cloud services. Due to be launched this week, IBM will be launching two main categories of service:
- IBM Smart Business: Software Development & Test
- IBM Smart Business: Virtual Desktops
The majority of IBM’s focus is in the private cloud enabling clients to, for example, own their own 42u cab complete with blade chassis and all the bells and whistles to run your own cloud. This package, selling for around $200k, allows IBM to charge for hardware rather than monthly utility based subscriptions similar to the majority of other players in the cloud arena.
The IBM cloud offering, historically unclear mainly due to the numerous points of entry and the possibility of alienating customers that could be in competition with them in the future, is now beginning to take shape.
Smart Business Development & Test
“IBM software delivery services for the cloud, enabling organizations to free up resources to gain rapid returns on their software investments by levering IBMï¿½s secure, scalable cloud delivery model for application lifecycle management”
This new service can either be run on IBM’s infrastructure or behind the client’s firewall. Alternatively hardware packages are available, allowing the client to essentially purchase an “out-the-box” cloud (named CloudBurst). A typical CloudBurst package is:
The basic CloudBurst package includes:
* 1 42U rack
* 1 BladeCenter Chassis
* 1 3650M2 Management Server, 8 cores, 24GB Ram
* 1 HS22 CloudBurst Management Blade, 8 cores, 48GB RAM
* 3 managed HS22 blades, 8 cores, 48GB RAM
* DS3400 FC attached storage
* IBM CloudBurst service management pack
* IBM Tivoli Provisioning Manager v7.1
* IBM Tivoli Monitoring v6.2.1
* IBM Systems Director 6.1.1 with Active Energy Manager; IBM ToolsCenter 1.0; IBM DS Storage Manager for DS4000 v10.36; LSI SMI-S provider for DS3400
* VMware VirtualCenter 2.5 U4; VMware ESXi 3.5 U4 hypervisor
IBM Smart Business: Virtual Desktops
“IBMï¿½s clients have also successfully leveraged cloud computing to virtualize desktops. Using up to 73 percent less power than traditional desktops and laptops, server-enabled virtualized desktops deliver a better end-user experience and can be more efficiently managed. Based on IBM internal data from client engagements, these solutions can also lower end-user IT support costs by up to 40 percent over traditional desktop environments.”
Again clients can decide whether they want to host this on IBM’s infrastructure or their own.
New to the forum? Take a moment to introduce yourself to everyone…
I’ll kick off:
I’m Paul, I graduated from University a few years ago and worked for a Financial Services company for a few years before leaving to become a partner in a web hosting company based in the UK. I have since spent the large part of my time researching the cloud from a business perspective rather (not technical!) and built this site as an information source for others who are also involved in the development of the cloud in one way or another.
Following their recent partnership announcement, Google and SalesForce have introduced Force.com for the Google App Engine. Web developers can now build web-based applications using the infrastructure from both companies.
I thought It’d be interesting to get everyone’s thoughts on what the most prevalent problems of cloud computing and cloud hosting are…
My two cents…I’m worried about vendor lock-in (as mentioned by GNU founder Richard-Stallman) – when you choose a cloud computing provider, such as gMail (for example), will I then be able to transfer my mail box to another provider at free-will? If not, will I be at the mercy of a price-hike for instance? I have seen the Open Cloud Manifesto, and I am indeed a supporter, but how “open” can you feasibly make the cloud?
I have a number of high traffic web sites. By high traffic I mean approximately 10,000 hits a day – so not high traffic in comparison to the big boys on the internet, but high enough to warrant a high powered VPS or a dedicated server.
There seems to be a lot of hype about the cloud and that it can offer huge cost savings. However, I gather that most clouds have a standard monthly charge on top of their utility billing so presumably traditional hosting is cheaper up until a point, but at that point economies of scale kicks in, the lines cross, and the cloud becomes the cheaper alternative.
Anyone know where that point is? My real question is, do you think I would make cost savings by hosting in the cloud or should I stick with my VPS for the time being? (my VPS is powerful and costs ï¿½50/month.
“I think everyone’s tired of the Cloud. I think, there is some Cloud fatigue going on. I think that people are starting to say — you are starting to see articles and say, this is all just a bunch of hype. In some respects, I think, it is. I think some people are attaching Cloud to their brand, into their services, where it doesn’t belong, but the basic fundamentals are really, really important. So I think it is absolutely reality and here is why.
It’s cheaper, better and more reliable. The last one, I think, I’ll get the most complaints about, but very rarely in business, do you get something as all three of these. There aren’t many advances in technology where you get things that are better cheaper and more reliable. You usually have to trade. This is why, this is such a buzzword, as you’re getting all three of these things. It really, really makes a difference for companies. So let’s talk about why it’s much cheaper.
So one thing is when you pool computing, you lower your CapEx. How you lower your CapEx? Well, you have higher utilization of your assets. This is the most talked about aspect of Cloud Computing, which is that most people have lots of infrastructure that is sitting unused or is sitting there waiting for when you might need it, but you can’t predict that. So that is a driving capital costs up for every single company that they don’t need it.
So, when you are sitting in a pool and you can take capacity as you need it, your CapEx goes much lower. Asset life, this is another piece of the capital equation is that when you’re sitting in a pool, it comes and goes, people can put applications on it. In traditional IT, often times what happens is you have hardware that sits with an application, and let’s say, you want to decommission that application.
Many times you’ll say, well, we need to do a hardware refresh and we need to get everything new, because we never know what’s going to happen. So assets end up getting not used for their useful life. When it’s pooled, it ends up being much easier to do that, which brings the cost down. So that’s one piece of dropping the cost. There is also the fact that it’s software-powered. That’s a huge difference; it is very, very automated.
You can go, sign up in our site at the rackspace cloud and you can launch a server, within five minutes, you’re going to have a root password and you can load code and you can get going or you can provision a domain and FTP, a WordPress blog and it will be on 100 servers. You can do this in five, ten minutes.
Now one person is involved, it’s completely automated, this is very, very low cost to do and really brings down operational cost. The other piece is you can pay for what you use, because you’re using software to carve up the infrastructure, you’ll have to pay for what you use and yet, you still have a big pool to call on if you need more.
The last piece, I think, which is not talked about a lot is that the delivered over the web piece also lowers cost because it brings focus to a company on things that are more important. No longer do you have to do the non-strategic things of running data centers and cabling racks or provisioning software, dealing with Dell, whatever it might be. You no longer have to do those pieces; you get a lot more focus in terms of the core pieces of your business.
Finally, by using a Cloud provider, you get economies of expertise. We’re a company of about 2700 people, we have experts in every single layer you can possibly think of. So networking, storage networks, load balancing, operating systems, security. We have sort of tons and tons of experts on these things and a lot of expertise is very expensive for any company to replicate. As technology gets more and more complicated, which it does everyday, trying to build the expertise around all of these different components gets very expensive for a company to do.
Unless technology and hosting is at the core of what you’re doing, you’re not going to want to do that. If you’re a retailer, you should not do that. If you’re a service provider, you should not do that. So, you get a lot of benefits by being able to use a third-party provider. So here’s an example, just a very class example. Let’s say, you’re someone who’s got a terabyte of videos that you want to serve over the web and you serve about ten terabytes a month of video content.
The traditional way to do it is to go get sort of an Enterprise SAN, call EMC or somebody, get Enterprise SAN, put in colocation, call a CDN provider, a Content Distribution provider or not. Just use traditional bandwidth; still going to be expensive when you’re buying in small quantities. $2000 a month for the storage, $8000 a month for the SAN.
What you could do with sort of our Cloud files offer today, it’s $0.15 a gig, so it’s a $150 a month and we have built in Limelight CDN, $0.22 a gig, so that’s going to be $2200 a month. So, 76% savings and you get the benefits of this because we’re a scale buyer, we have put a lot of engineering into this technology to make it cheap. This is the benefits of specialization.
So radically different price and huge savings. So this is a big reason you are seeing sort of the explosion of video over the web in this case study as it’s getting very cheap to do and very easy to do. So this is just one example. We can go through tons and tons of examples where you can save a ton of money, using the Cloud. So, why is it better? A couple of things to think about in terms of it being better. So, we’ve got this cheaper, how is it better? A couple of things; one is choice. Typically, in an IT department, you make platform decisions; you say, we are a .NET SQL shop or we are a LAMP shop or we are a Java Oracle shop.
Suddenly when you’re pulling, using service providers, you can choose a lot different technologies; you don’t have to be an expert in all of them. So, if the right tool is some Open Source project that’s out there and you want to play with it, you can provision a server on a Linux I/O shop and get it downloaded. So there is a lot of flexibility in terms of what you can use. You can mix and match and, you’re now is married to the skills you have internally.
Next one, scalability. So this is, I think, another way to tell that this is just speed. The ability to provision and get access to computing really, really quickly is tremendous and makes a big difference for companies and again, lowers the cost and get you to market really fast. The fact that you can really put up all your infrastructure in an hour is just — this used to take months to do. So, really makes a difference, both when you’re getting going and when you’re scaling and need to burst.
The last part is, it’s open. RightScale is a partner of ours, SOASTA, Cloudkick, these are people who are building tools on top of the public Clouds. So, if you want something that sort of makes monitoring really easier, provisioning really easy or clustering or load balancing, there are companies that are building tools on top of the Cloud providers that make all the common tasks you are doing all the time much easier.
So there is just so much power being added all the time. We have a company called Slicehost that it’s very similar to Cloud server’s offer and we had a developer, just randomly built an iPhone App that allows you to provision servers and launch servers and check the status of your servers, this was just done for us. Companies can go out there and use this iPhone App to run their infrastructure. This is the kind of thing that’s going to happen with all the large public Cloud providers out there, which makes them just tremendously powerful and more powerful everyday.
More reliable, this is the one that’s controversial because we’ve all seen the stories of outages, they make big headlines and everyone says how can you trust these Clouds? Look, I think there has been hiccups, there is going to continue to be some hiccups. What I would say is, and I stole this idea, someone at Google had made this analogy very, very similar to planes versus cars, which is when a plane — if something happens to a plane, it is major, major news and we all obsess on it and talk about it and what happened.
When someone dies in a car, it’s never discussed. This is very, very distributed. Which one is safer, I think that this is very, very similar to way the Clouds work. It’s very difficult to replicate the uptime that the leading Clouds have up there, but when they go down, the Internet goes down and it ends up being a massively noticeable event.
So, look, I think that for some time, this phenomenon is going to continue to slow the adoption, because the perception of quality is going to be a suspect. But I think overtime, you’re going to have fewer and fewer plane crashes and I think also people are going to start benchmarking their internal IT versus the services that they put out in the Cloud and they’re going to start making some rational choices around it.
So, I think that reliability really does matter. Okay, so this is why I think that the Cloud is real and it is not hype.”
Recently Google launched its Google Apps Outlook Sync stepping on the toes of Microsoft by luring their customers away from Microsoft Exchange email services.
Now it looks like Google may be doing something similar to Oracle with their new Fusion Tables service. Google released a post on the 9th June in its blog about a new SaaS database product designed specifically for “data management and collaboration: merging multiple data sources, discussion of the data, querying, visualization, and Web publishing”. The post stresses that the service is not “a traditional database system focusing on complicated SQL queries and transaction processing”.
The service already allows integration with the Google Visualisation API for a variety of ways to display data, extensive collaboration technology and data combining technology, with the promise of more functionality on its way.
Cost, manageability and the expected rising levels of spam in the region are the main reasons for academic institutions in East Africa to host their email in the cloud. Thousands of students, previously hosted on managed dedicated servers, have been migrated across to the Google’s Gmail service.
This is another example of the increasing trend to move services into the cloud. Should the trial be a success, expect more and more academic institutions to follow in their footsteps.
Avago and Genetech both made the switch from Microsoft Exchange to the new Google Apps Outlook Sync service. They explain the reasons why and how successful the changeover has been.
We have transcribed the video for users who don’t have audio. The actual video is at the bottom of the page
Robert J. Rudy:
Avago is a $1.7 billion semiconductor company. We decided to go to Google Apps, enterprise wide, 4100 users, the whole apps suite. So, my users are happy now but I also save $1.6 million a year.
Genentech was founded in 1976 and marked the beginning of the biotech revolution for healthcare.
We went with Google Apps because we see it as an innovative leader in cloud computing and bringing new technology to our users so that they can work more effectively and really focus on what matters most which is innovating healthcare.
Some of our users like Outlook, and so we wanted to be able to get the best of cloud computing and Google Apps but allow them to continue to use their familiar Outlook client.
Robert J. Rudy:
In the first week or so, after the big bang deployment, 39% of the users were still using Outlook because we had given them the choice to ease the change management angst. But in four weeks, it was down to 12%, so to be able to give them a full feature, high speed performance capability similar to the native web client, we heard about this from Google and asked to be an early adopter for it and then we implemented it.
Google Apps Sync for Microsoft Outlook gives them choice, makes it easy, and allows us to deliver an integrated solution to everyone.
Robert J. Rudy:
It eliminates the performance hit from using the IMAP for protocol. It gives them a richer experience than they had because theyï¿½re not waiting for things. They actually say itï¿½s like a miracle because they really just have the same environment they had before which is really what they wanted all along.
Google Apps Sync for Microsoft Outlook makes it easier than ever for enterprises to move to G Apps because they can. for those users that want to stay with the Outlook experience, they can do that and the enterprise can be free from much of the other infrastructure that you need for Microsoft Outlook support.
Robert J. Rudy:
The elimination of servers and storage, and backups and tapes, and people to manage all those activities are — Google Apps enterprise has actually cost one-sixth of what the previous incumbent Microsoft exchange environment was.Itï¿½s basically seamless with the Google Apps backend. Itï¿½s an amazing thing and I guess the thing that really popped out to me was how fast it runs.
Is Google really committed to the enterprise? Google Apps Sync for Microsoft Outlook shows the depth of their commitment and really bringing a terrific user experience and options for end users, while delivering the cost savings, the simplicity, the reliability, and the performance of cloud computing.
Robert J. Rudy:
So for CIOs who are considering moving to Google Apps Premier and who are concerned with the integration with Microsoft, I could tell them that this new plug-in eliminates that gap. The functionality is incredible, the user productivity is equivalent to the native web client, but the benefit for IT, managers and CIOs is that you also get the benefit from the huge cost savings from eliminating all the backend infrastructure associated with exchange.