back to article Internet, schminternet: Boffins propose private 100Gbps HYPERNET

A "hypernet" to be known as the Pacific Research Platform, shuffling data at up to 100Gb/s, will be established between US West Coast laboratories and several supercomputers thanks to a $5m grant from the National Science Foundation (NSF). The grant will span five years, reports the New York Times, and will connect the NSF's …

  1. Jim 59

    Smarr adds that one internet-connected server at the University of California recently received 35,000 false login attempts in a single day.

    To be fair, a single Wordpress server can get this from one attack in a few hours. It isn't a rarity. The Register probably get far more.

  2. Anonymous Coward
    Coat

    Darn....

    ....I'm going to have halt my roll out of MegaNet and start work on UltraSuperNet

    1. Steve Foster
      Joke

      Re: Darn....

      Bah, we're working on LudicrousNet here...

  3. Anonymous Coward
    Anonymous Coward

    Given the nature of the network, I would think one way they can increase security is to mandate some form of identity and authentication for every node in the network that's vetted before it's permitted to even work on it. IOW, unlike on the commercial Internet where anonymity is seen as a boon and therefore protected to some degree, you don't want anonymity on this new network, so if it's built from the ground up with a strong sense of node identity and attribution, then it will have an easier time fending off intruders and isolating problem nodes if one becomes compromised.

    1. Preston Munchensonton

      Pfft. The Internet includes enough to identify people as file sharers. That should be enough for most people (i.e. I trust no one but x.x.x.x -- rinse, repeat). Instead, what you're suggesting is harder to do, since most servers don't have user-level identities outside of node addresses or services. If you want to enforce this new level of identity on your megaultrasupernet, then just force all services to require TLS with certs for client and server.

      There's nothing new about this new NSF setup, outside of the speed. The airgap method has been around a long time and I doubt there will be much in terms of new network architectures, despite the claims from the blowhards involved.

      1. Anonymous Coward
        Anonymous Coward

        I'm saying they should go one better and use a new protocol that's built from the ground up with better attribution. Sorta like baking TLS into the protocol so that it's there from the beginning so that one couldn't be anonymous even if they tried (unlike with the Internet where a determined person could use proxies and so on to hide their identity). It would also help guard against attachment attacks since it would be an "alien" protocol to the TCP/IP of the Internet.

  4. Anonymous Coward
    Anonymous Coward

    We could have fasterthanthatnet here, only as BT have most of the fiber in the ground but won't sell dark fiber, they just gouge you to light it on your behalf and we only have standard Internet.

  5. Anonymous Coward
    Anonymous Coward

    PC as a bridge

    Being a university resource it would not surprise me if someone bridges it to the internet in a PC. Not high bandwidth but still a hole in the "air gap" security.

  6. Jon Massey
    Headmaster

    That's...

    not how you couple ST connectors!

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon