• On MovieTome: See the villain of IRON MAN 2!
January 11, 2009 7:01 PM PST

The cost of cloud adoption

by Dave Rosenberg

Most people assume that running applications in the cloud, and specifically on Amazon Elastic Compute Cloud, is automatically less expensive then running in your own data center.

The short answer is that the EC2 may not actually be the cheapest route, but it can provide faster time to market and additional revenue, even if it actually costs more to run.

I read a post on Geva Perry's Thinking Out Cloud blog Sunday that got me thinking about cloud economics, and if there is a missing link in the costs associated with cloud services and specifically what happens if you run everything you own on EC2 24 hours a day, seven days a week, 356 days a year.

Geva illustrates the point:

In other words, let's say you're buying a server from Dell for $5,800, and you expect to use that server for three years. You also have to pay a 10 percent annual maintenance fee on that server, so the total cost over three years is $7,540. Amazon, on the other hand, charges 40 cents per hour for an equivalent large instance on EC2. A simple calculation will show that using the EC2 instance for three years would cost you:

4 cents x 24 hours a day x 365 days a year x 3 years = $10,512

And that's excluding other charges from Amazon for bandwidth usage and other services, but it includes maintenance, which falls on Amazon. This also assumes that you are using the Amazon Machine Image 24-7-365, which may not be true, but we'll get to that later.

In essence, you are borrowing money from Amazon to buy a server, and paying the loan back to them in monthly payments of $292 ($10,512 over 36 months). Which means it's an annual interest rate of roughly 15 percent. Is that a good deal? Maybe.

As I explained in my $1 million example above, it all depends on your situation and alternatives: for example, can you invest that money is sales, marketing, R&D, and other activities that will produce more than a 15 percent annual return?

For many users, cost won't be the leading factor in cloud adoption (as witnessed in the math above). Control and security will trump everything else. And if you have just one company using the cloud, many other issues of multitenancy go away, reducing the time from development to production.

At this point, with Amazon as the hulking giant, the first company to implement a near-clone of EC2 and its associated components that you can deploy internally has a huge available market.

Dave Rosenberg dishes up "Software, Interrupted" with nearly 15 years of technology and marketing experience that spans from Bell Labs to multiple start-up IPOs to open-source enterprise software companies. He is co-founder of MuleSource and currently serves as the general manager of Hardy Way. He is a member of the CNET Blog Network and is not an employee of CNET. Disclosure. You can contact Dave via e-mail at softwareinterrupted@gmail.com.
Recent posts from Software, Interrupted
Ramen robots invade Japanese restaurant
Firefox 3.5 and the potential of Web typography
Blizzard chooses cloud over LAN for new game
Japan continues to build robot army
Ricoh jumps from copiers to the cloud
China bans online 'gold farming'
Japan airport starts motorized tricycle patrols
Why Oracle will continue to win
Add a Comment (Log in or register) (5 Comments)
  • prev
  • 1
  • next
by lmasanti January 12, 2009 7:50 AM PST
quote:
"...but it can provide faster time to market and additional revenue, even if it actually costs more to run."

All "subscriptiond" business models run in the same idea: cheap to enter, costly to be maintained,
Did the telcos invent it?
Reply to this comment
by kswartz26 January 12, 2009 10:18 PM PST
May want to check your math. $10512 is the equivalent of borrowing $7540 and paying *11%* interest compounded annually. Not 15%. Minor quibble, as your point still remains valid (perhaps more so).
Reply to this comment
by nh99 January 13, 2009 1:40 PM PST
The real benefit comes when you need to add servers. Compare the cost, time and technical skill required to add more (clustered) servers to handle your application when it gets popular. That's where EC2 really shines - you can pay for exactly the amount of servers needed to keep your application responsive. You can scale up for your peak demand, but you're not stuck paying for those servers later if demand drops back down. It enables you to create a business where costs and income move in sync.
Reply to this comment
by baetica January 16, 2009 10:39 AM PST
Your analysis is overly simplistic. Amazon also provides power, space, cooling and provides labor time to unpack, assemble rack, cable and power your server. They will also "replace" it instantaneouly if any components break. Over the five year period, these costs are easily a few thousand. Advantage AWS.

(I'd also dispute that the EC2 "large" instance is really equivalent to a $6k dell server -- it's got less processing power)
Reply to this comment
by ChrisFleck February 4, 2009 11:39 AM PST
Good points. Aside from the overhead costs of premise vs. cloud, the real savings come from the ability to turn off EC2 and not over buying capacity that sits idle. The right answer can often be " Premise Plus Cloud " as I point out on my post.

http://community.citrix.com/x/7AHDAg
Reply to this comment
(5 Comments)
  • prev
  • 1
  • next
advertisement
Click Here

Look before leaping to short URLs

Fueled by Twitter's rise, services that scrunch Web addresses are taking off. They bring a host of problems, but some are working to fix them.

In Utah desert, it's bombs away

road trip At the massive Utah Test & Training Range, the Air Force runs more than 15,000 sorties a year to ensure that pilots and weapons are on the mark.
• Photos: Training and testing

About Software, Interrupted

In "Software, Interrupted," Dave Rosenberg discusses disruption in the software market, as well as the products and services that keep business technology norms in perpetual flux.

With nearly 15 years of technology and marketing experience spanning from Bell Labs to multiple start-up IPOs, Dave co-founded open-source software company MuleSource and now serves as general manager of Hardy Way. He also happens to be a U.S. patent holder and a workaholic. Technology is his best friend and mortal enemy.

Add this feed to your online news reader

Software, Interrupted topics

advertisement
advertisement

Inside CNET News

Scroll Left Scroll Right