Stories
Slash Boxes
Comments

SoylentNews is people

posted by mrpg on Monday June 11, @07:34AM   Printer-friendly
from the with-NFS? dept.

Submitted via IRC for Fnord666

The National Science Foundation (NSF) is announcing a $1.8 million grant for the initial development of a data storage network over the next two years. A collaborative team will combine their expertise, facilities and research challenges to develop the Open Storage Network (OSN). OSN will enable academic researchers across the nation to work with and share their data more efficiently than ever before.

[...] NSF's investment in OSN builds on a seed grant by Schmidt Futures — a philanthropic initiative founded by former Google Chairman Eric Schmidt –to enable the data transfer systems for the new network. These systems are designed to be low-cost, high-throughput, large-capacity, and capable of matching the speed of a 100-gigabit network connection with only a small number of nodes. This configuration will help to ensure that OSN can eventually be deployed in many universities across the U.S. to leverage prior investments and establish sustainable management for the overall storage network.

Source: NSF Supports Development of New Nationwide Data Storage Network


Original Submission

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: -1, Spam) by Anonymous Coward on Monday June 11, @08:21AM

    by Anonymous Coward on Monday June 11, @08:21AM (#691314)

    "How dare you!"
    "No, how dare you! I can't believe you'd suggest doing such a disgusting thing!"

    Two men could be heard arguing in the park. Each of them were furious enough to scare away most onlookers, and so no one had stopped their heated argument quite yet. This continued for a few more minutes, but just when the argument was about to escalate into violence, a third man - who had been observing the situation all along - approached them and calmly said, "Now, now. Settle down, gentleman. I believe I have the perfect compromise that will satisfy everyone involved." Then, the man elucidated his idea; after that, it did not take long for the plan to be put into motion...

    Later, two men could be seen standing in the park near a large tree; both of them appeared to be in a state of ecstasy. One of the men had been directly involved in the previous argument, and the other man was the one who suggested the compromise. These two men were overlooking a magnificent scene with expressions full of delight. What caused them to be in such a state? Justice.

    It was true justice. A man - who had also been involved in the previous argument - could be seen hanging from the tree, lynched; his body was in tatters, suggesting that he had been tortured beforehand. Surrounding the tree were countless naked children; not a single one of them moved even so much as an inch. This was a scene that would undoubtedly bring joy to anyone's heart, and it was all thanks to the art of compromise.

    And what a fair, just, and wondrous compromise it was.

  • (Score: 2) by MichaelDavidCrawford on Monday June 11, @09:56AM (1 child)

    but to enable academics to use each others' computers without having to actually visit these computers in person.

    However that often required them to FTP huge data sets to those other computers. While far less costly than plane fare is was slow as well as a heavy drain on the ARPANET's scarce resources.

    Around the time I started at Tech in 1982, I read somewhere that there were "almost three hundred" hosts on the ARPANET.

    Will Wonders Never Cease?

    --
    My United States Social Security Number Is 518-92-8663
    • (Score: 2, Funny) by Muad'Dave on Monday June 11, @11:21AM

      by Muad'Dave (1413) on Monday June 11, @11:21AM (#691352)

      > FTP huge data sets

      ... datasets that could exceed ten million bytes in some extreme cases.

  • (Score: 3, Insightful) by Muad'Dave on Monday June 11, @11:35AM (1 child)

    by Muad'Dave (1413) on Monday June 11, @11:35AM (#691354)

    ... it's the metadata and indexing that's important.

    Without a well-defined way to represent the contents of the datasets and a way to effectively search them, all the data storage in the world is useless by anyone outside your group.

    If I have a simple dataset of time vs temperature for my area, the metadata for that information should include the axes of the data - in this case, time vs temperature, using standard units - (secs vs K for instance) and each axis should have more detail. For instance, the time axis should be described as completely as possible - time resolution, accuracy, step size, number of data points, time start and end (vs a standard epoch), etc. The temp axis should have similar detail - lowest temp, highest temp, accuracy, precision, etc. The location of the measurements should have similar metadata.

    This all supposes that there can be agreed-upon data format(s) and compression schemes - there are so many to choose from, and all have pluses and minuses.

    • (Score: 2) by MichaelDavidCrawford on Monday June 11, @06:19PM

      at least it intended to when I was in grad school.

      For a little while I was in a group at SLAC that was designing a C++ replacement for the byzantine CERNLIB. That replacement included a self-describing data format.

      But I got a little loopy and so dropped out of school. I didn't follow their efforts after that.

      --
      My United States Social Security Number Is 518-92-8663
(1)