Slashdot Mirror


What Kind of Data Center Can You Build With $500M?

coondoggie writes "So, if the government gave your company $500 million to spend on building a new data center what would you buy and how would you build it? Well, the Social Security Administration is about to find out. As part of the stimulus bill, or the American Recovery and Reinvestment Act of 2009, the SSA got the tidy little sum to replace its National Computer Center. The SSA in fact says it will need closer to $800 million to fund a new IT infrastructure, including the new data center — the physical building, power and cooling infrastructure, IT hardware, and systems applications. (This is addition to a $72 million backup facility currently under construction in Durham, North Carolina)."

18 of 204 comments (clear)

  1. The Unfortunate Reality of Maintaining Legacy by eldavojohn · · Score: 4, Insightful

    I will start with the assumption that this data center must be non-homogeneous. Get an assessment of all the projects that are using the current system you're going to replace (you know, the one with 36 million lines of COBOL code?). Because the number one priority of the customer (other projects) is going to be the lengthy transition from that to current technology. Prepare yourselves for this: Some of the projects aren't going to have any funding to do jackshit. Which means that the awesome spaghetti coded current system that's held together with COBOL duct tape needs to remain intact in some form. Not ideal situation but an uncomfortable truth. I'm thinking you would want to set aside 10% or $50 million or so for this (just throwing out a figure).

    --
    My work here is dung.
    1. Re:The Unfortunate Reality of Maintaining Legacy by Anonymous Coward · · Score: 5, Interesting

      As someone who maintains (part of) that 36 million lines of COBOL code, it's not as spaghetti as you might think. Many parts of it are quite clean, they just haven't been touched in years but they're simple and they work. The problem is, we're facing a demographic time bomb. The folks who wrote the system got hired in a huge spree in the 60s and 70s, and are reaching retirement age with full government pensions. Lots of them retire and come back as contractors, but we're still losing a lot of them.

    2. Re:The Unfortunate Reality of Maintaining Legacy by jellomizer · · Score: 5, Insightful

      I think more colleges as part of their Computer Science Elective they should be a class on Legacy System Maintenance. Legacy systems are not as bad as everyone makes them out to be. As well for the most part languages such as COBOL and FORTRAN are really not that hard to follow and learn. In many ways they are easier then the newer languages, as they are designed to do particular things and do them well, and not like C,C++,Java,.NET try to be the end all be all language. The old code is usually focused on the business logic while newer code seems to be working more with trying to get the formatting correct, and being well organized and modularized (which is a good thing too). But the old code you get PROCESS_CLAIM.CBL all the code that you need to process the claims are there, so when you fix the code there the problem is fixed. Having done a lot of work in FORTRAN myself I have found that there isn't that much spaghetti going on. Yes there is the GOTO statement but it is usually limited to ERROR cases where if something critical failes it GOTO ErrorNum and displays the error and quits the app.

      For these legacy apps normally when something needs to be changed is because there was a change in the business process not because of a bug.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    3. Re:The Unfortunate Reality of Maintaining Legacy by SatanicPuppy · · Score: 4, Insightful

      As someone who maintains COBOL driven legacy systems, I disagree. Why teach that stuff in college? Those jobs are much more rare than other programming gigs, and they tend to be held by lifers (I inhereted mine when 2 people hit retirement age, and the third decided she didn't want to do it alone). If you DO get one of those jobs, the learning code for the obscure hand-coded systems is going to be vastly higher than the language.

      COBOL isn't that hard to pick up. Maintaining legacy crap code is the same whether it's VB or COBOL or RPG, and the vast vast majority of your headaches will come from environmental quirks (old school databases with fixed width data, packed binary decimal numbers that no one uses anymore, etc).

      The biggest problem with COBOL in the modern world is that its designed in reverse. It treats CPU cycles and RAM like they're the most precious things on earth, so a program will make live changes as it goes along (to conserve RAM and minimize disk IO), and is designed to fail in a dirty state (in the middle of everything, so you can't re-run it) on the chance that it'll preserve cycles. It's a real maintenance headache.

      --
      ad logicam Claiming a proposition is false because it was presented as the conclusion of a fallacious argument.
    4. Re:The Unfortunate Reality of Maintaining Legacy by thePowerOfGrayskull · · Score: 4, Insightful

      Everybody* else knew the language was dead.**

      *everybody: the disturbingly large subset of programmers who are unaware that COBOL and FORTRAN still power the vast majority of the financial and healthcare infrastructures of the world.

      **dead: working quietly in the background without need of constant updates and maintenance

    5. Re:The Unfortunate Reality of Maintaining Legacy by bdenton42 · · Score: 4, Insightful

      There is nothing wrong with a functional 20 year out of date COBOL application. It is a bigger waste of company/governmental resources to rewrite it just because you want to use the language flavor of the month.

      So they should have rewritten it in 'C' 20 years ago, then of course you have to go OO and rewrite it to C++ 15 years ago, then Java was really cool so 10 years ago rewrite again, and now M$ has taken over the world so we better port it to .NET today, but you've just wasted 4 projects to simply get the existing functionality and your company is out of business (or added 0.0001% more to the national debt if you are the government).

      There are very good reasons to port old projects but just doing it because it's 20 years out of date is not one of them.

  2. Mimicking Private Industry? by eldavojohn · · Score: 4, Insightful

    Could you approach Google and ask them to license their ideas on server and data 'pod' technology for your sharded databases? I'm not saying build the whole thing like this but with $500 million, you could probably have a large section to search and sharded databases that mimics Google. I don't think there's anything wrong with following the leader in that department. This probably isn't the best solution for relational databases so I would think another architecture would be in place for your MySQL and Postgres traditional database layouts. And that would be just huge centralized servers running virtualized instances of Linux with MySQL or Postgres.

    --
    My work here is dung.
  3. Why build one... by Bandman · · Score: 4, Interesting

    ...when you could have two for twice as much?

    OK, actually two 250 million dollar datacenters, but I love that quote from Contact.

    Why tie up that much money in one site when you could build two world class structures and have full redundancy?

    1. Re:Why build one... by Anonymous Coward · · Score: 5, Interesting

      I concur. There are 3 empty data centers along the I-81 corridor (within 2-3 hours of DC) that stand empty, reminders of the dot com bust. Why build new when you can retrofit an existing center.

    2. Re:Why build one... by timelorde · · Score: 5, Insightful

      Because they're not in my Congressional district?

  4. How about a location first by schwit1 · · Score: 4, Insightful

    Putting the operation in a location that is cost effective would make the taxpayers very happy. The DC area is too expensive. Maybe an old missile facility in Wyoming or the Dakotas.

    1. Re:How about a location first by duckintheface · · Score: 4, Interesting

      Look at the locations where Google is putting it's new data centers. They are in rural parts of the Carolinas on property adjacent to power plants. Google negotiates uninterruptable power contracts in fast growing states on cheap land.

      --
      "He took a duck in the face at 250 knots." -- William Gibson, Pattern Recognition
  5. Something modest by IWantMoreSpamPlease · · Score: 4, Funny

    Then spend the rest on hookers and blow.

    Hey, the gov't does that all the time, why are you staring at me like that??

    --
    So rise up, all ye lost ones, as one, we'll claw the clouds.
  6. ..near an industry that can use hot water. by paul.schulz · · Score: 4, Interesting

    Find an industry that would otherwise need a gas fired boiler and on-sell the heat.

    Other options:
    - Heated public swimming pool
    - Source of community/public heating

  7. What Kind of Data Center Can You Build With $500M? by Rik+Sweeney · · Score: 5, Funny

    Data Center: $10M*

    Hookers and Beer: $490M

    *I reckon I could get that down to $6M by cutting corners though, so that's another $4M for the beer :)

  8. Simple by Opportunist · · Score: 4, Funny

    200m for me, 200m for you, 100m for the guy in India who builds a data center.

    --
    We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
  9. I don't know... by Cornwallis · · Score: 5, Insightful

    ...but I'll bet you $500 million that: 1) It won't be nearly enough money; 2) It will be obsolete before it is finished; 3) It won't be finished before Social Security runs completely out of money (which will coincide nicely with my scheduled retirement); 4) [Fill in the blank]

  10. Why do they need a data center that large? by ciellarg · · Score: 4, Funny

    Congress has already bankrupted the SSA fund, so all they need is one computeer that can add and subtract "0" very quickly....unless they need something that can divide by zero :)