4 reasons why cloud computing is efficient

44 Comments

There have been a few recent analyses showing that cloud computing has significant efficiency and cost advantages. The most recent one with which I am directly familiar was conducted by Jon Taylor’s team at WSP Environment & Energy for Salesforce.com, and it showed per-transaction emissions reductions averaging 95 percent for companies that shift to using the cloud.

I can think of four reasons why cloud computing is (with few exceptions) significantly more energy efficient than using in-house data centers:

1. Economies of scale. It’s cheaper for bigger cloud computing folks to make efficiency improvements because they can spread the costs over a larger server base and can afford to have more dedicated folks focused on efficiency improvements.

For example, there are usually significant fixed costs of implementing simple techniques to improve Power Usage Effectiveness (PUE), like the costs of doing an equipment inventory and assessment of data center airflow (same for implementing institutional changes like charging users per kW instead of per square foot of floor area). Whenever there are costs that are substantially fixed (i.e. only weakly related to the size of the facility), bigger operations have an advantage because they can spread the costs over more transactions, equipment, or floor area.

There’s also a substantial advantage to having “in house” expertise devoted to efficiency, instead of having staff split between different jobs. Technology changes so rapidly that it’s hard for people not devoted to efficiency to keep up as well as those that are.

2. Diversity and aggregation. More users, more diverse users, and more users in different places means computing loads are spread over the day, allowing for increased equipment utilization. Typical in house data centers have server utilizations of 5-15 percent and sometimes much less, whereas cloud facilities for major vendors are more in the 30-40 percent range.

3. Flexibility. Cloud installations use virtualization and other techniques to separate the software from the characteristics of physical servers (some call this “abstraction of physical from virtual layers”).  This sounds like a great thing for software and total costs, but why is it an energy issue?

Using this technique means that you can redesign servers to optimize them and drop certain energy costly features. For example, if software can route around physical servers that die, you no longer need to have two power supplies in each server; the death of any one particular server doesn’t matter to the delivery of IT services.

In essence, this technique redefines the concept of reliability from one that is based on the reliability of a particular piece of hardware to one that is based on the reliability of the delivery of the IT services of interest, and this is a much more sensible approach.

4. Ability to sidestep organizational issues instead of having to address them head-on (which is hard and slow). While most company in-house IT operations face the problem of a disconnect between IT departments driving server purchases, and facilities departments paying the electric bill, that problem has largely been solved for the cloud providers. They generally have one data center budget and clear responsibilities assigned to one person with decision making authority.

Economies of scale are more powerful in the cloud scenario, because you’ve gotten rid of the impediments to taking action and can allow those economies to work their magic. Finally, it’s much easier and cheaper for people stuck with the in-house organizations to use a credit card to buy cloud services instead of waiting around for their internal IT organization to get its act together.

These four big energy advantages will over time translate into more and more pressure for companies to adopt cloud services, because the economic advantages (driven by the energy advantages) are so large.  And it’s not just energy costs, it’s the capital cost of all the supporting equipment, which in a standard in-house facility can be $25,000/kW and (together with the energy costs) add up to half or more of the total costs of the facility (for details see Koomey, Jonathan G., Christian Belady, Michael Patterson, Anthony Santos, and Klaus-Dieter Lange. 2009. Assessing trends over time in performance, costs, and energy use for servers. Oakland, CA: Analytics Press.  August 17.)

Of course, there are still issues to work out. For example, people haven’t really ironed out the complexities about liability for cloud outages.  And there will always be providers who will want to have their own in-house facilities for security reasons (like big financial institutions). But even in that case, the benefits of a virtualized cloud infrastructure can be brought to the in-house facilities.

You won’t get the same diversity, but the other benefits of cloud will still be powerful. I’ve also heard of companies creating “private clouds” for use by other companies that pay in to use them on a “members-only” basis, thus dealing with the diversity and security issues. So things are evolving rapidly, but the economic benefits are so large that we’ll see a whole lot more cloud computing in coming years.

This post originally appeared on Jonathan Koomey’s blog.

Jonathan Koomey is a researcher, author, lecturer, and entrepreneur whose work spans climate solutions, critical thinking skills, and the energy and environmental effects of information technology. He’s been a Consulting Professor at Stanford University since 2004, and recently held visiting professorships at Yale (Fall 2009) and Stanford (2003–4 and Fall 2008), and worked as a researcher and scientist at Lawrence Berkeley National Laboratory (LBNL) for more than two decades

Image courtesy of Pike Research, Grigorieff Photography, GigaOM Events

44 Comments

Hitesh

Folks can I make this below mentioned anology:

Cloud Computing Vs In house Data Centres are going to coexist just like Hotmail Vs Inhouse eMail exchange system within the organisations.

John Harrington, Jr.

All of your points are dead on, and there are companies out there proving them for you. Fiberlink’s cloud-based MDM platform lowers costs and risks with faster results, gives immediate insight and control, latest product versions without manual updates, and much faster time-to-value than any on-premises solution (with devices enrolled in minutes!)

Check out this webinar case study presented by GSA to get some great insight: http://bit.ly/pdtN0V

Joseph Starwood

Your observations on side-stepping in-house departmental boundary problems are unique and helpful. The head-butting that accompanies infrastructure changes is often noisy, distracting, and time-consuming. Side-stepping is certainly a valid approach and benefit.
I’d also like to mention improved focus on infrastructure and facility decisions as another cloud benefit. Cloud providers are pressed to drive results to their bottom line. This helps them focus on getting the appropriate products and configurations. This is not always true for in-house IT departments. I once watched an infrastructure team spend 3 months pouring over hunting magazines and archery catalogs in preparation for the upcoming deer season. During those months, they spent 15 minutes in one meeting deciding the number, model, and configuration of new servers for a mission-critical application.
The bottom-line brings things into focus. Cloud providers have this going for them.
__ Joseph Starwood

Steve Bowes-Phipps

Let’s take this argument back to whether or not this article is correct in its analysis.

Unfortunately, I believe it is flawed in just about every point that’s been made.

1. Our organisation has proven that you don’t need to have a big data centre to be efficient. We’ve won awards to prove it.
2. I would have thought that having ‘diverse users’ would be an argument against cloud not for it – how can the architecture of cloud possibly run diverse applications efficiently?
3. Believe it or not, all other organisations can also purchase and use virtualisation in their data centres – this is not a technology restricted to cloud companies.
4. I am not even slightly convinced that cloud providers are able to solve the issue of “Why innovate on energy efficiency, when I don’t pay the bill? connundrum. I took a tour round a very large cloud infrastructure provider and they leave their facility lights around the cavernous and long corridors and in their even more cavernous data rooms on constantly – burning kWh as though PUE didn’t exist. When I inquired as to why, they just shrugged their shoulders and said that their customers prefer the areas to be lit up.

If you’re a Data Centre Manager who doesn’t care about wasting your organisation’s money by failing to act to reduce the TCO of your data centres then you should be fired and replaced by someone who does.

Alpha Solutions

Cloud computing is the way of the future, regardless if the general public likes it or not. It has been on it’s way for quite some time and the heavyweights are pushing it even harder. Google +1 is an effort to entice more people into Google Apps. Microsoft’s “Live” and new Office 365 is their foot in the door.
What all of us that live outside of cyberspace need to remember is that we cannot fully depend on cloud technology. There is something called “backup” that will save us all from the clouds when the wind blows too hard.

Hans

EXECUTIVE SUMMARY: Cloud computing is hot. But severe business risks and challenges are involved to retrieve the promised business advantages of cloud computing. Conclusions: The main disadvantages are: security, loss of control, unstable cost structure , potentially decreased business flexibility, integration problems. An adaptive architecture, including a new business model must be developed to retrieve business value Organizational changes are needed for adapting the technology (procedures, skills, business model) Formulate and maintain an exit procedure to change from CSP (Cloud Service Provider). Cloud computing technologies and business models have not yet reached maturity. Standards are being developed.

kmashraf

http://bit.ly/nfXjNA
This article describes one of the major pitfalls for companies from the rest of the world that use cloud services provided by vendors situated in the US of A

jan klamer

The biggest con against cloud computing; not everywhere and always there is infrastructure!

Was in south america recently – only 20% of the time i could get a net connection that was very poor. I was SO happy that 100% of the time I just carried all personal data along in a tiny HDD – not bigger than my smartphone :-)) no clouds, i prefer the sun and 100% efficiency!

Phil

I believe in the cloud model/benefits. That said for large enterprises “private clouds” owned and operated by said enterprises will most likely be the way forward. Why large enterprises need anything more then SLA’s for MAN/WAN/ISP services is a mystery to me. I guess it goes back to the vertical market they operate in and the sensitivity of the “data” they use/create/manipulate.

Dee

If I write my own scientific-engineering software and would like to upload ONLY the resultant excutable programs in a Cloud system, how do, or can I, protect both the executable and source codes against any unauthorized applications and/or downloading? The idea is that instead of “selling” the software in a form of CD to prospective licensees or authorized users, let them ONLY access the executable codes or programs but can run the programs either on the users’ machines not connected to the Cloud or my machines connected to the cloud? Please kindly enlighten me, anyone. Thx

Asad

I think cloud computing can be a cost saving one, but for now let us all consider to use for the all non company secret computing. I think every company should have their own server as back up security precaution until cloud service provides improves the details of security, dedication, etc. Let us face the reality, we all want to save money, jobs, and productivity, but what we don’t want is cloud to take the road our all manufactory jobs took which is “Out sourcing” jobs to others counties.

Mark S

Not sure I buy into all this rather academic blathering. Sorry. Sort of. I do however plan on naming my next child “Cloud”. Stinks if it’s a girl but I get a new ‘Vette with all the money that gets magically mailed to me.

denis

Do I think that the cloud is the way of the future ? Damn straight.
Do I consider the cloud as part of my strategic vision ? Absolutely.

Are we there yet ? Not even on my “production” roadmap. Leaving aside that point 3 has nothing to do with clouds, If I can’t quickly and easily reconfigure the way I want it, then as experience has shown, it is quicker and cheaper to do it in house. I certainly challenge the cloud marketing hype of “time to market” as a valid indicator. It should be, and will be, but at the moment I can deliver quicker “in house”.

Charanjeev

I believe the answer lies in industry focussed clouds (eg. A highly secure and available financial services cloud, vs. A less stringent sales and logistics cloud). Another moot point is the risk assessment raised by someone earlier, if the cloud fails to live up to expectation, what kind of a denial of service is your business capable of absorbing and for how long. It may be prudent to engage in an analysis of what parts of the business can be clouded, what can be clouded with in-house DR and what must remain strictly in-house. Cloud computing is a worthy opportunity but not a totalitarian solution like most other technological developments. The onus remains with the not yet extinct IT director’s office to plan a vision and factor in cloud computing. IMHO.

Kat

In the early days when ‘cloud’ wasn’t even called Cloud working in the ISP and very early elearning space we talked to the big vendors – and the determination was to get the user locked into purchasing applications, that were at the beck and call of the suppliers – to drive up the cost of applications. Rather then that one off purchase with an upgrade maybe 2-3 years later, you pay for an upgrade at their call. In fact you have to be connected to do business. The emphasis wasn’t for the benefit of the user.

Francis

Something worth reading: http://www.pcworld.com/businesscenter/article/227189/moving_to_the_cloud_isnt_for_everyone_what_to_consider_first.html

In short, the Cloud is not necessarily for everyone… OK I didn’t invent anything here, it’s in the Title of the article…

Planning is the key.
I also believe critical data should remain in house (that one is only a personal opinion).

In the end it’s a Risk assessment exercise: how much confidence are you prepared to give (and effectively pay for) to your “Cloud Provider” ?

Vic

Totally agree with Francis on this point. Who and where will your company data be stored in the world. Also cloud computing is still emerging with unknown risks involved so can the cloud prover ensure the safety of the data they store as a service to their customers. Current charges for the services is not exactly within affordable means and the risk involved is not exactly every it companies cup of tea. Data withheld or lost ultimately results in business failure.

Alok TIwari

Lot of buzz words around the topic.
My questions If any body wishes to answer??
What’s about security? In Last 2 years Two major players faced the issue.
configuration ? Unable to get server details from cloud hosting providers
Max computing cycles supported? This is most unanswered questions from cloud hosting provider
Does your Cloud Hosting Company supports SLA
ping me if somebody have answers.

MeerkatMac

Cloud computing isn’t new, just the name is. It was called time-shared computing in the 60’s and early 70’s when companies bought time on mainframe computers to do their computational work. The mini-computer, and then personal computers sprang up to address the problems of security, performance, and availability by recognizing computing capabilities as a strategic resource and putting control and direct access to them in the hands of the companies and individuals that needed them.

Now the industry is trying to come full circle and tell us sharing computer resources is really a great idea. Yeah, great idea for those wanting you to have to come to them for all of your computational needs, so they can take control of your future and charge you metered rates to stay in the game.

This is outsourcing plain and simple. And history shows that outsourcing not only outsources your internal costs, but also your strategic advantage.

Rob Greenberg

Sure much of this is outsourcing, but not many companies get their competitive advantage from IT. Outsourcing IT through “the cloud” allows companies to focus on their true competitive advantages.

Sam

We recently finished an exhaustive comparison of what it costs to run our IT department vs going to the cloud. The cost difference was staggering, the cheapest cloud offering (signing a 3 year lease) was 6x as expensive as it was for us to run it in house (this includes staffing colo rent, ISPs, software licensing), and that doesn’t even factor in the additional services provided by the IT department like monitoring (which costs extra in the cloud) and deep dive troubleshooting (which is largely unavailable).

asetip

I believed in cloud computing will saved cost in germ of energy and resources. But whatever it is, life cycle analysis should be done first. Due to different nature of opertion, not all organisations could enjoy the saving. It is depends on how intensive and frequent changes in reprograming the apps that suit the business operation vs productivities.

Zoltan

It is just a game change. You may forget about some of the earlier issues when joining a Cloud environment while having some new issues (SAN / Disaster Recovery issues mostly out while problems with services in case of a bill dispute are in). I believe larger, professional Service Providers may work at an economy of scale you never have in smaller datacenters. It is just true for all parts including energy use. Security, Integration – some new sort of challenges.

Hans

I agree with Ethereal;

If you believe the hype, IT departments are about to become redundant and all that shiny hardware and flashing lights in the server room are about to become a thing of the past. We can look forward to the history and archaeology museums setting up displays of the IT Director’s office next door to the caveman exhibit!

Well, sorry to burst the cloud computing marketing bubble, but this is just not so. Cloud computing is full of issues as the technology develops, and as with any relatively new and innovative service or product, the application to the real world is uncovering new issues just as much as discovering new ways in which old issues, such as privacy have to be handled. Cloud computing offers some extremely excellent benefits but, developers and users must exercise a degree of circumspection and engage in regular reality checks!

David Spratt

I heard people talk like you in the 1980’s when the PC and the client server networking model emerged. I suggest you look harder at this before you go the way the COBOL programmers and Data Processing Departments did back then.

Sharique Abdullah

Any concept that evolves from the scratch has issues, and that’s what evolution is for. Cloud computing would eventually grow into something better than what it currently is. Technology is meant to be like that.

EtherealMind

Blah Blah Cloud.

Again.

Clouds are for poor people who can’t afford to build a proper application. When you get on public transport that is shared with other people, it never works quite the way you want. And there is no competitive advantage.

Clouds are going to take a small, but significant group of companies that can’t afford decent computing power. And will live with the limitations. None of this is ever considered by the Cloud boosters. More reality, less “blah blah”.

Jenny Levy

Wow. I think you are clearly unaware of who is successfully using cloud computing already. They definitely aren’t companies “that can’t afford decent computing power.” They are companies like Twitter, Salesforce, Cisco

David Spratt

You are correct about one thing. Only rich people can afford products like Microsoft Outlook and CRM. These are expensive and inflexible apps that are hard to configure and use – gmail and salesforce.com, by comparison provide real competitive advantage on the basis of scale, ease of use and price. I think its you who needs to tone down the uninformed blah blah bla

Jubal

The primary reason I see cloud computing as potentially more efficient is simple. If a new “greener” technology comes out, it is far easier for Amazon or Microsoft to implement that in 10 large datacenter vs. asking every IT team with a server room to update their infrastructure.

Eric

Cloud computing is a very broad term so it’s always best to be as specific as possible when talking about Cloud. Microsoft 365 for instance is very different than saying hosting with the IBM Smart Cloud… Cloud Computing will continue to grow however, there is still a lot of growth within the Cloud Providers themselves in order for the economies of scale to be fully realized. Most of our customers are still very concerned about security and sla agreements.

David Spratt

This “define cloud computing” is a tired old hack usually trotted out by people trapped in the spotlights of this oncoming train. Cloud computing is simple – it is a service accessible via a web browser and a network connection – nothing more needed by way of definition.

Sharique Abdullah

Cloud is NOT a service accessible via a web browser, it is much more than that. Cloud is simply a more evolved form of previously known concept of web/server hosting services. It is just that the hosting concept is now grown to embrace other services as well, be it an ERP, a banking service, ATMs or pretty much anything that would require technology (more precisely Information Technology) to back it.

Wayne Barber

David is absolutely right, We own a private data centre and we are finding that cloud based solutions are more accessable than ever, you dont need expensive lease lines anymore to connect to cloud based data – we are able to bond broadband with great effect. Smartphones now provide great ‘fall-back’ for email collection and you dont have to go full measure when choosing hosted.

Ray

IMO Cloud computing is a marketing concept. Technology makes a difference marketing concepts don’t. The technology that drives cloud will remain the marketing concept will wither and die.

Mike

I have one question… when company A gets a bill from cloud computing company and they say… we aren’t paying for that. We disagree with your how you are billing us… and cloud computing company says well you’re not getting your email then! Or whatever application they want to host… what does company A do? Their Business Suffers? Over a billing dispute. There are a few “concerns” about cloud computing however for a small company I think it makes sense.

Lenny

You can’t just stop sending a client’s email due to a bill dispute. In the UK at least, this is illegal.

Ed

Not necessarily It would depend on the dispute resolution terms that are in the SLA or contract you signed.

Comments are closed.