Blog

Infrastructure-as-a-Service Breakdown

There’s an interesting post by Advanced Systems Group that provides the top-level costs they were able to derive from an internal study they did to compare public and private cloud implementations.  It’s the start of a good discussion (questions/comments I posted are still in moderation).

Their post compares three different cloud options from a cost perspective: 1) Public Cloud from Amazon EC2; 2) Purchased Private Cloud infrastructure; and 3) Leased private cloud infrastructure.

All three systems were based around the equivalent of (using AWS parlance) 10 Extra Large CPU instances, 60 Large CPU instances, and 30 Small CPU instances.

The cost calculations were run by ASG over 3 years assuming an increase of 15% in compute demand, and not accounting for depreciation.

When ASG noted in an answer to a comment that all instances are assumed to be running 24/7 my curiosity was piqued to drill in a little bit.  The problem is, solely comparing private and public based upon 100% utilization removes one of the most compelling reason for the cloud - a pay-as-you-use model.  It skews the conclusion from the start.  (How many data centers run at 100% utilization anyway?)

Also, from a rough calculation, we can see that their Amazon numbers appeared to be based upon the cost of On-Demand Instances.  24/7 utilization over a three year period is not a cost effective use model for the On-Demand pricing model.

Performing a straight hardware costs calculation, (I’m assuming that ASG were adding in some additional costs for storage and data transfers to get to their top line numbers, as I couldn’t get there from a simple calculation.  Their 3Yr cumulative cost for the Public Cloud scenario was slightly over $1M, where as mine below came in slightly under), I got the following top line numbers for an On-Demand instance scenario (the least economical).  For completeness, I also added in the two more sensible scenarios of either a rolling 1 Year Reserved Instance model or a one-time 3 Year Instance model.

Public vs Private Cloud Costs (100% Utilization)

And so, taking these basic calculations in isolation of all other considerations, we can see that from a purely hardware resource expenditure perspective, a Public Cloud implementation is roughly the same or even less than a private cloud implementation.

However, this is still only part of the story.  As mentioned, a key draw to public cloud computing is the elasticity of the model - Pay-as-you-Use.  And so, considering a selection of rough percentage utilizations, we get the following breakdown of the different hardware costs of a public cloud implementation.

AWS Public Cloud Costs: Time vs Model vs Utilization

The comparison relative to utilization becomes even more stark when charted.

Cloud Cost Comparison vs Percentage Utilization

Not only is AWS more economical over utilization rates over extended periods, but it provides flexibility.  For example, if the software used on the servers is upgraded (as is likely) and hardware requirements increase (equally likely, given history), moving an image (AMI) from a Small to a Large instance is straight forward (for an “IT guy”).  No hardware needs to be thrown out or shipped in.

Additionally, there is some flexibility around migrating across business models.  Of course, it is easier to upgrade than downgrade

Posted under cloud

The Economist this week published an article (Tanks in the Cloud) that talked about an attempt to fill in the cloud market knowledge gap left by industry analysts such as Forrester Research.  The problem the analysts have is that the further down you go in the Cloud stack, the less forthcoming companies are in divulging their revenue numbers.  Companies say even less about the number of servers they have in their data centers (which is where the tanks come in).

The 2010 revenue numbers of interest (see illustration) are SaaS (Software-as-a-Service) $11.7 Billion (Forrester Research); PaaS (Platform-as-a-Service) $311 Million (Forrester Research); IaaS (Infrastructure-as-a-Service) ~$1 Billion.

Cloud Layers with 2010 Revenues

The “$1 Billion” figure for IaaS comes from an extrapolation from estimates of revenues generated for Amazon by its AWS (Amazon Web Services) business.  Two estimates were used, one performed by Randy Bias of Cloudscaling ($500M to $700M), and the other by UBS ($500M in 2010, and $750M in 2011 - a tidy growth rate).

And the tanks? I’ll leave it to the article, but the method applied by Cloudkick (acquired by Rackspace) and Guy Rosen estimated Amazon to be deploying around 90,000 virtual servers per day in their East Coast region alone.  The estimate is a little on shaky ground, but “in the land of the blind, the one eyed man is king” right?

Anyway, that’s a lot of servers.  It makes you think about software vendors considering turning their data centers into clouds to make their software available in a SaaS model.  The private-cloud-made-public is a good way to start at this nascent stage of market development, but how does this scale as we project out?

Posted under cloud

(It’s been too long since my last post (we’ve been *really* busy at Xuropa), but this is too big…)

There has been buzz about a Google OS for a while, but it’s now official - Google is getting into the OS business.

It’s been something that has been begging to be done for years - a fast, light, and secure OS that gets you onto the web “immediately”.  Back in 2001I even batted around the idea (Linux with an Opera browser) at Merinta (R.I.P.)  as a way to turn that company around (not enough gas).  In ‘95 Ellison had the idea of the Network Computer and announced it at Comdex of that year.  This was while I was at National Semiconductor working on x86-based set-top-boxes and the beginnings of what became to be Internet Appliances.  As with everything - the first goes around never make it, especially when they’re this revolutionary.  The first attempts, ideas, and runs at the problem spur the industry, the environment, the infrastructure, and the thinking of the day to lay the ground work for “the one” to make it happen.  All the pieces are in place for this to be a great success.  Exciting times.

I particularly appreciate in the Google blog post the statement that the OS is a “natural extension of” the browser.  The reversal of the exact position Microsoft took in defending its bundling of the browser with Windows when it first released IE to kill Netscape.  Sweet irony.

But to my point about it being bigger that it initially appears.  For example, Michael Arrington’s post on TechCrunch misses the mark.  He states that “The browser is the platform.  The browser is the UI.”  He’s only half right and is really missing the bigger picture.  The correct interpretation of the announcement, IMO, is “The Cloud is the Platform.  The browser is the UI”.  And it’s not just Microsoft that should be worried, but Amazon, all other cloud computing providers, and subsequently telecoms companies.

I expect to see a plethora of cloud-based services and a Chrome-based API to access them to come along with the Chrome OS.  The first obvious service is storage, but in terms of applications the sky’s the limit.  In fact, combining the OS with the browser not only gives Google a competitive position against Microsoft, it significantly strengthens their position against Amazon and other Cloud vendors.  It’s another choke hold on access to services that will be too sweet for developers to pass up.  Before you know, there will be services and applications built upon the Chrome OS/Browser API’s and a whole infrastructure put in place before Microsoft and Amazon can say “Department of Justice”.

I’ve always said that people were spending too much time talking about Microsoft’s monopolistic position and not looking at Google enough.  But just like the advent of other standard platforms (eg. the IBM PC and Windows), this is going to open up tremendous opportunity for industry growth…and mass schizophrenia.

Posted under business, industry