• 0 Posts
  • 17 Comments
Joined 1 year ago
cake
Cake day: June 18th, 2023

help-circle


  • “Gamers Nexus, on the other hand, thinks the issue is more deep rooted and originates from a foundry-level fault.”

    • The GN piece makes it very clear that this claim is not definitely true but is a line of inquiry.
    • Intel statement does not definitely exclude this hypothesis, the flawed CPU might need the lower voltage to work around the flaw.
    • The obvious question this article does not address is what will be the performance hit for the patched parts?

    That’s a bit annoying to see GN so grossly misquoted when Steve spends half the run time of the video explaining that they are not sure of anything at this point.





  • If both networks 10.100.100.0/24? And 10.20.20.0/24 share the same level 2 Ethernet segment/vlan/broadcast domain, you don’t even need the third nic, you can setup a secondary IPv4 address on the private nic on the 10.20.20.0/24 network.

    I would not call that best practice, but if the number of host on the network is reasonable and you are aware of the security problems created, there’s nothing really wrong with this setup.

    Having two nics on the same Ethernet network is actually trickier since you have to do ARP filtering.


  • I have no experience about what you are trying to achieve, but rdma and related technologies (infiniband, qlogic, sr-iov, ROCE) is not it. These are network technologies that permit high bandwidth/low latency data transfer between hosts. Most of these bypass the IP stack entirely.

    Infiniband is a network stack that enable RDMA, it’s only vendor is now NVIDIA which acquired mellanox. Qlogic was another vendor, but it got acquired by Intel that tried to market it as Omnipath, but it was spinned off to Cornelis network.

    Sr-iov is a way to share an infiniband card to a virtual machine on the same host.

    ROCE is an implementation of the rdma software stack over ethernet instead of infiniband.




  • I’m not sure about the point of your setup but let’s ignore that :

    • Create a vlan “servers” on router B, assign a port (WITHOUT vlan tagging) to this vlan, patch a cable between this port and any port on rouer A
    • Put a static IP address 0.x on router B in this vlan
    • Enable routing between the default vlan and vlan “servers” on router B .
    • Configure router A to not distribute this IP address (by setting up a permanent DHCP lease for example)
    • On all your servers put a static route that says : “192.168.20.0/24 via 192.168.0.x”
    • If you can setup this route on A, things connected to A will work whether they have the route or not (it’s not a big deal but the routing would be assymetrical)

    If you can create a VLAN and a route on A, you can create a distinct “interconnect” VLAN and make all of that nice and clean without the extra static routes on the servers.