Please create an account to participate in the Slashdot moderation system

 


Forgot your password?
Close
typodupeerror
×
Science

NIST Studies Virus, DDoS Effect On Grids 54

Posted by timothy from the also-bad dept.
Ben writes "Scientists at the National Institute of Standards and Technology (NIST) have launched a new campaign to study the effect of viruses and denial of services attacks on grid computing systems. Specifically, they're developing models to establish vulnerability and find ways of fixing problems. But a grid's very strength -- its distibuted nature -- makes it vulnerable, indeed, they're finding. (Via Science Blog)"
This discussion has been archived. No new comments can be posted.

NIST Studies Virus, DDoS Effect On Grids More

NIST Studies Virus, DDoS Effect On Grids

Comments Filter:

  • Isolation (Score:5, Insightful)

    by Anonymous Coward on Monday August 02, 2004 @01:07AM (#9861576)
    Hmm, I've never seen any serious computing grids that are open to outside networks or that run windows.

    I think its common wisdom to isolate grids from the internet and other potentially hostile networks.
    • Good luck finding a network that's free of that virus, however. The kinds of people so clueless as to not understand the root cause of DoS attacks demand it because they are ignorant of alternatives and usually don't want to hear about them.

      I think its common wisdom to isolate grids from the internet and other potentially hostile networks.

      So you want to build a third internet? Sounds great, so long as you let me into internet 2, which was built specifically for grid computing on university sites. When

    • Umm you're only partially correct, viruses are ok, but its not just Windows that is vulenarable to DDoS attacks.

      Anyways, it doesn't matter since most grids that do any kind of important work are not open to internet.
      • Umm you're only partially correct, viruses are ok, but its not just Windows that is vulenarable to DDoS attacks.

        True, but it's the number of compromised Windows machines that allow for large DDos attacks in the first place. Get rid of those (if that's from better security practices or eliminating them) and you cripple DDoS attackers.

  • Sloppy sentence structure. (Score:4, Insightful)

    by SeaFox ( 739806 ) on Monday August 02, 2004 @01:09AM (#9861582)
    But a grid's very strength -- its distibuted nature -- makes it vulnerable, indeed, they're finding.

    Ewwww, awkward!

    Better: But the grid's very strength, its distributed nature, makes it more vulnerable to these types of attacks.

    Or: However, they're finding the grid's strength -- its distributed nature -- makes it vulnerable.

    This is one of those times dramatizing a sentence makes it worse.
    • But they're finding a grid's very strength -- its distributed nature -- makes it vulnerable, indeed.

      This is more fun than the Soul Train Scramble Board!

  • hah (Score:4, Funny)

    by andreyw ( 798182 ) on Monday August 02, 2004 @01:13AM (#9861593) Homepage
    Cue in jokes about studying the slashdot effect on grid computer...

    But seriously, what can simulate a DDoS more accurately than a bazillion slashdotters trying to RTFA and get the slashdot fix?
  • Did anyone else read that as "NIST Studies Virus, DDoS Effect on Girls?"

    I suppose DDoS'ing a girl is a pretty good way to give her a virus.

  • Article on NIST (Score:5, Informative)

    by Datasage ( 214357 ) * <DatasageNO@SPAMtheworldisgrey.com> on Monday August 02, 2004 @01:15AM (#9861598) Homepage Journal
    http://www.nist.gov/public_affairs/techbeat/tb2004 _0730.htm#tackling

    Does anyone know anything about the scienceblog.com site? It doesnt seem very reputable considering all the ads and the very obvious SEO technquies (look for the hidden links at the bottom).

    • Re:Article on NIST (Score:4, Insightful)

      by Anonymous Coward on Monday August 02, 2004 @04:13AM (#9861775)
      "Grid computing takes advantage of "down time" when computers are not using their full processing power to provide quick answers to problems in fields such as genomics, engineering design and financial services."

      NIST jumping on the Grid bandwagon without knowing
      what the term means. It's not about spare cycles;
      it's about doing reliable distributed computing
      in a loosely connected internet world. Most
      machines "on the Grid" are large, dedicated
      parallel farms bought for the purpose.

      Foster's most recent Grid definition is at:
      http://www.gridtoday.com/02/0722/100136.html [gridtoday.com]

      AM
      • Informative.

        A GRID is a non-centralized distributed system, sharing storage, processing and connectivity with quality of service guarantees, over open protocolls. Is that about right?

        Last summer, I was in Barcelona for a class trip. We visited the European Center for Parallellism of Barcelona [cepba.upc.es] (Like there are redundant arrays of Barcelonas. Well, there's the smaller town/city of Badelona right outside Barcelona.)

        Anyway...

        The CEPBA is a partnership with IBM. I talked to one of the IBM researchers there. T
    • tested link [nist.gov]. Yours had a problem besides having to cut and paste.

      Kind of a waste because it has exactly the same text as scienceblog. same text as the scienecblog.

  • But... (Score:4, Interesting)

    by xSquaredAdmin ( 725927 ) on Monday August 02, 2004 @01:18AM (#9861606)
    one of the disadvantages is that the same data will be sent to multiple computers. You can see that they used a grid computing system to create the article:

    From the first paragraph:
    Grid computing takes advantage of ''down time'' when computers are not using their full processing power to provide quick answers to problems in fields such as genomics, engineering design and financial services. While parallel processing typically involves tying together multiple computers at a single site--all using one piece of software--a computer grid may be much more geographically dispersed, composed of many heterogeneous computers whose availability may change over time.

    The third paragraph:
    Grid computing takes advantage of ''down time'' when computers are not using their full processing power to provide quick answers to problems in fields such as genomics, engineering design and financial services. While parallel processing typically involves tying together multiple computers at a single site--all using one piece of software--a computer grid may be much more geographically dispersed, composed of many heterogeneous computers whose availability may change over time.
    • Uh... Looks like a top summary to me.
    • That's kind of why I hate press releases. Well not only because news sites just distill them into articles without investigation, but because they contain information not relevant to the newsbit, but to make it more adworthy.

      For example, the release concerning Grand Haven's wireless network had information on Ottawa Wireless. The discription of Ottawa Wireless happens to say they also do WiMax, then says WiMax is a trademarked name. The problem is that Grand Haven's network didn't use it.

  • Freaking Classic! (Score:1, Funny)

    by Anonymous Coward
    This has nothing to do with the article, but of what I saw on the site while reading the article. To the left of the article is an ad for a t-shirt that has the caption "ASTHMA IS SEXY!", then the logo image, and below it, it says: "You knew it all along. Now own the shirt! Great gift for the lunger in your life." Heh, even the site selling the t-shirt is comical; check it out... Tee [tshirthell.com]

    The perfect web ad for the slashdot crowd!

  • (GNC) Grid's Not Cluster (Score:2, Informative)

    by Anonymous Coward
    Grid computing takes advantage of ''down time'' when computers are not using their full processing power to provide quick answers to problems in fields such as genomics, engineering design and financial services.

    The way I understand it, they're talking about all the cubies in Charles-Schwab and Raython and Citibank, with Folding@Home and Seti clients running as screensavers. Not dedicated clusters; just plain old ordinary workstations.

    And I'd imagine the viruii and worms and DNS outtages all take a s
  • This hypothesis seems to go against logic but maybe it holds some water until things change a little. When you have one very good point of control to access to the system where is the difference then? Also when distributed systems use multiple locations as well as multiple configurations doesn't this create a more survivable system? The laws of nature clearly say that biodiversity is directly linked to an ecosystem's success. An attack or technical problem can not effect every part of a diverse system in th
  • I've always just likened a grid to one big computer composed of many parts, which makes the next bit easier to visualise.

    As a grid basically has all the connected boxes on the same framework, the possible attacks could either be external or internal to the grid. If external then it's either the fault of the underlying OS or the grid itself and should of course be repaired. Just like a single computer.

    However I think it's unlikely that internal infection could occur (which is how most networks get infected

Slashdot Top Deals

The truth of a proposition has nothing to do with its credibility. And vice versa.

Close