After Amazon, how many clouds do we need?

20 Comments

With news that Google(s goog) and Microsoft(s msft) plan to take on the Amazon Web Services(s amzn) monolith with infrastructure services of their own, you have to ask: How many clouds do we need?

This Google-Microsoft news, broken this week by Derrick Harris,  proves to anyone who didn’t already realize it that Amazon is the biggest cloud computing force (by far) and, as such, wears a big fat target on its back. With the success of Amazon cloud services, which started out as plain vanilla infrastructure but has evolved to include workflow and storage gateways to enterprise data centers, Amazon’s got everyone — including big enterprise players like Microsoft, IBM(s ibm) and HP(s hp) — worried. Very worried.

These vendors are betting big that they can give Amazon a run for its money and that their cloud services will help them retain existing customers and (knock wood) win some newbies. Microsoft built Azure as a full-fledged platform as a service, but in the face of Amazon’s success had to tack to offer IaaS-type services, including VM Roles, which has been in beta for more than a year.

Amazon as enterprise apps platform? Don’t laugh

Take the news late this week that IBM is working with Ogilvy and Mather to move the advertising giant’s SAP(s sap) implementation from its current hosted environment to  “SmartCloud for SAP Applications hosted in IBM’s state-of-the-art, green Smarter Data Center.” (Note to IBM: brevity is beauty when it comes to branding.)

Don’t think that little tidbit is unrelated to last week’s announcement that SAP and Amazon together certified yet another SAP application — All-in-One — to run on Amazon’s EC2.  This sort of news validates Amazon as an enterprise-class cloud platform, and that’s the last thing IBM or HP or Microsoft wants to see happen.  So every one of these players — plus Google — are taking aim at Amazon.

Some hardware players, including HP, which is reportedly about to cut 30,000 jobs, see the cloud as a way to stay relevant, and oh, by the way, keep customers workloads running on their hardware and software.  HP’s OpenStack-based public cloud went to public beta earlier this month.

Case in point: Along with the SAP migration news,  IBM also said its SmartCloud Enterprise+, IBM’s managed enterprise cloud infrastructure offers:

“unprecedented support for both x86 and P-Series [servers] running … Windows, Linux and AIX on top of either VMware or PowerVM hypervisors….

and

SCE+ is designed to support different workloads and associated technology platforms including a new System z shared environment that will be available in the U.S. and U.K. later this year.

Hmmm. P-Series and System Z —  not exactly the sort of commodity hardware that modern webscale cloud companies run, but they are integral to IBM’s well-being.

Vendor clouds to lock customers in

This illustrates what prospective buyers should know: Despite all the talk about openness and interoperability, a vendor’s cloud will be that vendor’s cloud. It represents a way to make sure customers run that company’s hardware and software as long as possible. But legacy IT vendors are not alone in trying to keep customers on the farm.

Amazon is making its own offerings stickier so that the more higher-value services a customer uses, the harder it will be to move to another cloud.  As Amazon continues what one competitor calls its “Sherman’s march on Atlanta,” legacy IT vendors are building cloud services as fast as they can in hopes that they can keep their customers in-house. For them, there had better be demand for at least one more cloud.

There will doubtless be more  discussion on this and other cloud topics at the GigaOM Structure conference in San Francisco next month.

Photo courtesy of Shutterstock user Ohmega1982,

20 Comments

Gregg McKnight

How many clouds do we need? To get a scope of the question I would look at the existing baseline of computation. There are approximately 10 million x86 servers sold each year and the average server life cycle is about 3.5 years. Thus at any time there are at least 35 million x86 servers that represent the possible computation workload that one could consider migrating to “the cloud”. I have not seen public information documenting the number of servers that Amazon hosts in its public cloud but I would guess the number is somewhere between 500,000 to 750,000 with many of these servers used for its own retail operations. Anecdotal evidence suggests that Google owns on the order of 1.5M servers but the majority of these are for search operations. A reasonable guess might be that Google, Amazon and others currently have a total of at most about 1M to 2M servers specifically dedicated to cloud computing (non-search) workloads. So it seems to me that we have just scratched the surface for the total capacity potential of public clouds.

CloudSigma

Great article. It’s a good question to ask – whether there is room enough in the market for another slew of big cloud vendors offering “business as usual” IaaS. At CloudSigma, we agree that it’s going to be difficult for behemoths like Google and Microsoft to square off against Amazon. However, we do think there is plenty of room for competition from newer providers who can offer more flexible, transparent and cost-effective clouds.

For instance, our IaaS platform provides a more scalable cloud, with a true utility computing, that doesn’t lock customers into restrictive proprietary frameworks. Our cloud is built to allow customers to build full virtual data centers in the cloud. It’s this sort of innovation that will ultimately erode AWS market share.

Daniel Steeves

Reblogged this on DanielSteeves and commented:
Om, I can’t believe you publish this drivel.

Yes, sure let’s have one cloud and maybe only one computer provider: monopolies are so good for the world!

What a ridiculous view… tripe!

Daniel Steeves

Om, I can’t believe you publish this drivel.

Yes, sure let’s have one cloud and maybe only one computer provider: monopolies are so good for the world!

What a ridiculous view… tripe!

Ezana Berhane

I don’t get why people hate on the cloud, as if it has to be an all or nothing option. Why does everything have to either be local or cloud? Why can’t it be both? They complement each other very well. In this case, I see Amazon as the clear winner because they have already established market share in various sectors. I only see Google or Microsoft having a chance if they build something innovative. Google has a chance, Microsoft…fuggedaboutit.

scott herson

It is going to be a very hard shift (culturally) for HP and Microsoft to get into the cloud hosting game. I do not believe that developers/start-ups (who are embracing cloud the most) will be easily convinced by the likes HP/Microsoft. The existing enterprise customer bases of HP, Microsoft and IBM are very slow movers to the multi-tenancy of the cloud because of control, jobs, compliance, security, etc. The real winner will be a company that can satisfy both the developer/start-up sectors as well as the enterprise… I think Google has the best likelihood of pulling that off. Amazon didn’t plan for this sort of success in the cloud and will doubtfully ever win the enterprise – they are busy selling books and doing content deals (amazon studios etc). If I were Bezo’s, I would spin Amazon off into its own gig. There will likely be a lot of IaaS names within the Gartner Magic Quadrant that get folded into these big firms by way of acquisition – let them continue the innovation for now. Finally, this is like Data Center Colo-Cation in 1997… its just getting started.

Eddie

What might be really interesting to watch, regarding Google, is if they were to do something antithesis of AWS, more in the spirit of what Vint Cerf has blogged about (regarding no lock in and intra-cloud communications)! Mr. Cerf is a Google VP an Chief Internet Evangelist, and he did more for the Internet than Al Gore ever did :-)

deeceefar2

It is obvious the future of all computing is cloud based. It simply doesn’t make sense for home users and average office workers to continue to have terminals that do the computing when the cloud can do it faster for less money and deliver it where ever you go. So there is massive growth left for the sector and as the market gets bigger interoperability is going to become important, and the prices are going to get driven lower. Right now you see that the average server processor in the cloud does not differ to significantly from home computers, but as the cloud because the main market for processors the processing architectures are going to go through massive changes to optimize around specific workloads. We are in such the infancy of the cloud there is HUGE opportunity left to innovate in that space and really the only thing holding back massive growth is large scale internet cloud, enterprise cloud interoperability.

Daniel Steeves

Infancy? We’ve been selling cloud in the form of iaas and saas for a very long time.

And no, frankly, *all* computing will not be in the cloud. Much will but never all – and never major enterprise data.

Funny enough huge amounts of that major enterprise stuff is in a cloud and has been for years – but we are talking a series of private, secured clouds…

As for ‘everything’ in the cloud, what happens when the network drops (could be a bit of road digging and BOOM your non-diverse but very cheap connectivity to your simple, cheap but probably non-redundant cloud? or what happens when the power goes out?

For me and for my clients and their clients, we all keep computing.

Dave Mackey

I’ll be happy if we have three major players in the field – e.g. Amazon, Microsoft, and Google. There are tons more players…but if we can have a few out front that push each other to stay competitive on price and innovate on features…that’ll work.

Rahul

That’s like saying how many OS or laptop models do we need, or why isn’t everyone just happy with an iPad. Different needs will be catered to by different vendors, an enterprise worried about security will prefer a private cloud, a small and medium business might prefer AWS, a start up might…… And so on.

Jeff Schneider

Hey Barb – I’m guessing that less than 1% of enterprise computing is currently done in the cloud. Said another way, the remaining 99% of the spend is up for grabs… I think there is plenty of room for everyone to make a nice living.

Ahmad

Amazon cloud price is still high… We need another big player that can offer a cheap price.

Keith Townsend

I actually hoped that Microsoft was going to originally offer IaaS solution instead of Azure. IaaS made much more since than Azure at the time. Even still, I think if anyone can give AWS a run it will be a Microsoft Hyper-V native cloud.

For all Windows based enterprises, a Microsoft IaaS offering will give pause to the thought of hosting your Windows workloads on AWS. I think OpenStack may have more to worry about than AWS especially, if Microsoft chooses to allow mobility between their cloud and others based on a native Hyper-V virtual machine.

Eddie

One of the fellows at Microsoft who is all over their cloud strategy is Rolf Harms (Director Corporate Strategy). Check out his cloud economics papers etc. (btw he used to work at SAP).

tim

The cloud sucks….its unsecure, if your interbet goes down and you need your data, you are sol. Physical data is by far more reliable

Comments are closed.