[nickysielicki]

The calculation under “Quiz 2: GPU nodes“ is incorrect, to the best of my knowledge. There aren’t enough ports for each GPU and/or for each switch (less the crossbar connections) to fully realize the 450GB/s that’s theoretically possible, which is why 3.2TB/s of internode bandwidth is what’s offered on all of the major cloud providers and the reference systems. If it was 3.6TB/s, this would produce internode bottlenecks in any distributed ring workload.

Shamelessly: I’m open to work if anyone is hiring.

[aschleck]

It's been a while since I thought about this but isn't the reason providers advertise only 3.2tbps because that's the limit of a single node's connection to the IB network? DGX is spec'ed to pair each H100 with a Connect-X 7 NIC and those cap out at 400gbps. 8 gpus * 400gbps / gpu = 3.2tbps.

Quiz 2 is confusingly worded but is, iiuc, referring to intranode GPU connections rather than internode networking.

[nickysielicki]

Try running a single node all to all with sharp disabled. I don’t believe you’ll see 450GB/s.

[charleshn]

Yes, 450GB/s is the per GPU bandwidth in the nvlink domain. 3.2Tbps is the per-host bandwidth in the scale out IB/Ethernet domain.

[jacobaustin123]

I believe this is correct. For an H100, the 4 NVLink switches each have 64 ports supporting 25GB/s each, and each GPU uses a total of 18 ports. This gives us 450GB/s bandwidth within the node. But once you start trying to leave the node, you're limited by the per-node InfiniBand cabling, which only gives you 400GB/s out of the entire node (50GB / GPU).

[xtacy]

Is it GBps (gigabytes per second) or Gbps (giga bits per second)? I see mixed usage in this comment thread so I’m left wondering what it actually is.

The article is consistent and uses Gigabytes.

[shaklee3]

GBps