Google Ads

Enter your email address:

Delivered by FeedBurner

This form does not yet contain any fields.

    Microsoft writes humorous blog post to educate SW developers of the power costs to run their code

    Microsoft has a blog on "How Much Does Your Code Cost?", interjecting humor into a typically dry topic.

    The big difference is that with cloud computing, you’re renting computing power in a data center somewhere. As far as you’re concerned, it could be on Saturn. Except that the latency figures might be a bit excessive. If you’ve accidentally opened one of those magazines your network administrator takes with him to the bathroom, you might know that these data centers contain racks and racks of servers, all with lots of twinkling lights. If you’ve ever been to a data center, you’ll know that they can be very hot near the server fans, much colder around the cooling vents, and noisy everywhere. All this activity results from removing the heat that the servers produce. But that heat doesn’t get there all by itself – the servers create it from the electricity they use. What’s more, it requires even more electricity to remove that heat.

    Consider sending this post on to those who are involved in SW decisions to get them thinking of the impact of their SW code.

    When you’re up against deadlines to turn in a software project, you probably are focused on ensuring that you meet the functionality requirements set out in the design specification. If you have enough time, you might consider trying to maximize performance. You might also try to document your code thoroughly so that anyone taking over the project doesn’t need to run Windows for Telepaths to work out what your subroutines actually do. But there is probably one area that you don’t consider: the cost of your code.

    You mean what it costs to write the code, right? No.

    Er, how about what it costs to compile? You’re getting warmer...

    What it costs to support? No, colder again.

    OK, you win. What costs do you mean?

    I mean what it costs to run your code. In the bad old days, when clouds were just white fluffy things in the sky and all applications ran on real hardware in a server room somewhere or on users’ PCs, then cost simply wasn’t a factor. Sure, you might generate more costs if your application needed beefier hardware to run, but that came out of the cable-pluggers’ capital budget, and we all know that computer hardware needs changing every other year, so the bean-counters didn’t twig. A survey by Avanade showed that 50% of IT departments don’t even budget for the cost of electricity to run their IT systems. For more information, see this Avanade News Release, at

    Life would be so much easier in the data center if SW developers and others thought of the data center infrastructure costs direct relationship to the code they write and how it is architected.

    The good thing is cloud computing is helping to get SW developers to think about the costs to run their code.

    If you deploy applications into the cloud, it is highly likely that your service provider will be charging you based on the energy that you use. Although you don’t see electricity itemized as kw/hr, you are billed for CPU, RAM, storage and network resources, all of which consume electricity. The more powerful processor with more memory costs more, not just because the cost of the components, but because they consume more electricity. In many ways, this is an excellent business model, as you don’t have to buy the hardware, maintain it, depreciate it, and finally, replace it. You simply pay for what you use. Or putting it another way, you pay for the resources you use. And this is the point at which you need to ask yourself: How much does my code cost? When power usage directly affects the cost of running your applications, a power-efficient program is more likely to be a profitable one.

    The blog post references Visual Studio and Intel resources to help SW developers.

    It is possible that future versions of Visual Studio will include options for checking your code for power usage. Until that time, following these recommendations should help minimize the running costs of your applications within a cloud-based environment.

    1. Reduce or eliminate accesses to the hard disk. Use buffering or batch up I/O requests.
    2. Do not use timers and polling to check for process completion. Each time the application polls, it wakes up the processor. Use event triggering to notify completion of a process instead.
    3. Make intelligent use of multiple threads to reduce computation times, but do not generate threads that the application cannot use effectively.
    4. With multiple threads, ensure the threads are balanced and one is not taking all the resources.
    5. Monitor carefully for memory leaks and free up unused memory.
    6. Use additional tools to identify and profile power usage.

    For more ideas on how to reduce the memory, check out the following resources and tools:

    Energy-Efficient Software Checklist, at

    Creating Energy-Efficient Software, at

    Intel PowerInformer, at

    Application Energy Toolkit, at

    Click to read more ...


    With recent volcanic activity how long will it take before anyone considers Iceland for a data center?

    We've all seen the Iceland gov't delegation at various data center conferences, and they have had a web site up to promote their site.


    Our competitive operating environment compares favourably with leading countries in the industrial world. Add to this our green renewable energy, low tax structure, high education levels, abundant land and competitive costs for skilled labour, and you will quickly discover that Iceland is a strong candidate as a location for international data centre operations.

    Connected to the world

    Just when they thought the gov't instability and Fiber connectivity were addressed, their statement on low risk natural disasters is irrelevant.

    MSNBC reports on the latest volcanic activity and flooding.

    Iceland evacuates hundreds as volcano erupts again


    Associated Press Writers, Associated Press Writers

    updated 33 minutes ago

    REYKJAVIK, Iceland - A volcano under a glacier in Iceland erupted Wednesday for the second time in less than a month, melting ice, shooting smoke and steam into the air, closing a major road and forcing hundreds of people to flee rising floodwaters.

    Authorities evacuated 800 residents from around the Eyjafjallajokull glacier as rivers rose by up to 10 feet (3 meters).

    Emergency officials and scientists said the eruption under the ice cap was 10 to 20 times more powerful than one last month, and carried a much greater risk of widespread flooding.

    Can anyone be taken seriously now if they listed Iceland as a site for a data center?

    The risk may have been no higher for Iceland than other countries, but perception of risks given recent volcanic activity have changed.  The general public would say the risk of Volcanic activity is 100%. :-)

    Click to read more ...


    Google's Eric Schmidt discusses Sharing and Mobile strategy

    GigaOm analyzes this video of Eric Schmidt at Atmosphere.

    And throws this out as summary of key issues Eric presents.

    Schmidt made two specific comments about resource allocation, saying that the hardest and most pressing engineering issues facing Google today are around sharing and mobile. He was talking to the enterprise execs present but his statements were so absolute I think it’s fair to apply them more broadly.

    “Companies are about sharing,” Schmidt said. “One of the new things in the last five years about the web is that it enables sharing-sensitive apps.” He continued,

    I think of calendars as incredibly boring, but I’m wrong, calendars are incredibly interesting because they’re incredibly shared. So from a computer science perspective, all of a sudden we have our top engineers who want to build calendars. I’m going, what’s wrong with you guys? But in fact it’s a very interesting example. Spreadsheets are similar, the most interesting spreadsheets are highly, highly interlinked, something I didn’t know, and was not possible with the previous technology — Microsoft technology made it very difficult because they were not built in that model.

    Google's Don Dodge (recently laid off by Microsoft) adds his perspective on the threat to Office.

    Erick Schonfeld at Techcrunch says; "Slowly but surely, Google keeps trying to chip away at Microsoft’s core Office productivity suite with Google Docs, its free online word processor, spreadsheet, and presentation software. Today, Google Drawing is being added to the mix and Google Docs and Spreadsheets is getting a major realtime update."

    David Berlind at InformationWeek is much more aggressive. "Make no mistake about it. Google is going for Microsoft's jugular. The deathmatch is on and, at the very least, it's for bragging rights to what we at InformationWeek are calling the "collaborative backbone." It becomes a battle that's less about Google Docs versus Microsoft Office and much more about the collaborative infrastructure behind Google Apps versus Microsoft's SharePoint and Exchange."

    And provides a graph to illustrate his point.

    This competitive positioning chart illustrates where Google is coming from, and where it hopes to go in the future. It is the classic Innovators Dilemma competitive curve. Time will tell how it shakes out. The move to the cloud seems to be pretty clear. Only the slope of the curve and speed seems to be in question.


    And, let's not forget the changes from Mobile.

    As the mobile Internet becomes central for both consumer and corporate users, the core product questions are interoperability, security and safety, Schmidt said. “What’s important is to get the mobile experience right, because mobility will ultimately be the way you provision most of your services,” he added, saying that Google considers phones, tablets and netbooks mobile experiences.

    These are all things we are thinking about as we get the GreenM3 NPO rolling, and how we will approach data center information sharing.  In some ways you could contrast what we are thinking of in an Open and Transparent approach to data center innovation vs. the status quo.  It is close to the comparison of Microsoft's individual authoring thinking vs. Google's team collaboration.

    Click to read more ...


    ASHRAE Standard 90.1 Data Center Documents

    After writing on Google's post regarding ASHRAE's standard 90.1 and requirement for economizer and talking to Google's Chris Malone, I decided to find the documents.

    Here is the proposed Addendum bu.

    6.5.1 Economizers. Each cooling system that has a fan shall include either an air orwater economizer meeting the requirements of Sections through

    There are a through k in just section 6.5.1

    j. Systems primarily serving computer rooms where
    1) the total design cooling load of all computer rooms in the building is less than 3,000,000 Btu/h (880,000 kW) and the building in which they are located is not served by a centralized chilled water plant, or

    And Addendum cy

    What is hard to figure out is why ASHRAE is making economizers a requirement for cooling systems instead of a performance based solution, but if you meet an efficiency improvement in table 6.3.2 you are allowed you to eliminate economizers.


    Does addendum cy supersede addendum bu?  addendum cy doesn't have the above section j.

    DataCenterKnowledge references addendum bu.

    ASHRAE, for its part, says it welcomes the feedback on the proposed changes. “ASHRAE is committed to excellence in the consensus standard development process and encourages anyone with comments regarding the proposed addendum regarding data centers (addendum bu) to participate in the public review process,” it said.

    Are you confused?  I am.

    Click to read more ...


    Can Data Centers benefit from Supply Chain Management concepts?

    Currently, I am studying data center site selection, and have been asking the question what is wrong with data centers having 1% of the cost being in the land when other commercial real estate will typically have land 20-25% of the cost.  One big thing most miss is land is not a cost, it is a non-depreciable asset. 

    Capital assets that are inexhaustible or where the useful life does not diminish or expire over time, such as land and land improvements. Infrastructure assets reported using the modified approach to depreciation are also not depreciated.

    Land is not an expense, it is an investment.  So, land should be looked evaluated on its ROI, not it's overall cost, including land improvements. 

    Which then led me to think why is it data centers don't use more supply chain management concepts which would address issues like land cost in the overall solution and most likely save you much more than the cost of the land?

    Supply Chain Management is defined as.

    Supply chain management (SCM) is the management of a network of interconnected businesses involved in the ultimate provision of product and service packages required by end customers (Harland, 1996).[1] Supply Chain Management spans all movement and storage of raw materials, work-in-process inventory, and finished goods from point of origin to point of consumption (supply chain).

    Another definition is provided by the APICS Dictionary when it defines SCM as the "design, planning, execution, control, and monitoring of supply chain activities with the objective of creating net value, building a competitive infrastructure, leveraging worldwide logistics, synchronizing supply with demand, and measuring performance globally."

    Can't you think of all the different groups and vendors involved in providing data center and IT services as a supply chain management problem?  Is the CIO in charge of the supply chain? Maybe.

    Here is a piece of irony from a article on supply chain management.  Supply chain management SW is a mess.

    Supply chain management software is possibly the most fractured group of software applications on the planet. Each of the five major supply chain steps previously outlined is comprised of dozens of specific tasks, many of which have their own specific software. Some vendors have assembled many of these different chunks of software together under a single roof, but no one has a complete package that is right for every company. For example, most companies need to track demand, supply, manufacturing status, logistics (i.e. where things are in the supply chain), and distribution. They also need to share data with supply chain partners at an ever increasing rate. While products from large ERP vendors like SAP's Advanced Planner and Optimizer (APO) can perform many or all of these tasks, because each industry's supply chain has a unique set of challenges, many companies decide to go with targeted best of breed products instead, even if some integration is an inevitable consequence.

    So, if a bunch of people who focus only on supply chain management can't get the software right, how can the data center industry get the right software to run data centers like a supply chain?

    I think I have an answer on how to approach supply chain management for data centers.  The first step is to identify the problem, then test what approaches solve the problem best. The fragmentation and silos is the opportunity to address.  How do you pull all the pieces together?  My ideas are based on using social networking and memetics.

    More to come.

    Click to read more ...