Gigabyte B343-C40-AAJ1 Block Diagram
Here is the block diagram for the node:
The block diagram is interesting because there are two PCIe Gen5 x4 lanes going to the 2.5″ bays. There are also two SATA III lanes from the chipset to the 2.5″ side. Another interesting takeaway is the PCIe Gen3 x1 to the M.2 slot. This is really for boot media. We can also see the edge connector board being connected with PCIe Gen4 x4.
Gigabyte B343-C40-AAJ1 Management
On the management side, we get fairly standard web management. That is less interesting. What is more interesting is how these are connected.
Instead of needing one management port per server, the chassis management function allows for the nodes to be managed from a single interface. You will see two ports on the back, and that is for another interesting feature.
Gigabyte has the ability to use these ports to daisy chain. That means you can pack these 3U 10-node chassis into a rack and then daisy chain those together. That means many of these chassis can be serviced by a single management network port.
This may seem like a small feature, but instead of needing an entire management switch for 9-12U of systems, one only needs a single port on a switch.
Gigabyte B343-C40-AAJ1 Performance
On the performance side, we have gone over this a few times already during the past few months. Namely in our AMD EPYC 4005 Grado is Great and Intel is Exposed piece.
These days, the performance of the AMD EPYC 4005 series is really good. Up to sixteen Zen 5 cores at high clock speeds really help with performance.
For the lower-end hosting segment, the AMD EPYC 4005 allows for more performance per node versus previous generation multi-node designs. That means either using higher-power 16-core parts for more performance or using lower-power designs for lower costs.
One valid take would be why not just use a dual socket Turin Dense server and have even more cores per U. On the other hand, the dedicate hosting market is always one where customers want their own servers. That is really the point of this type of node.
I can totally see a server like this being filled with 9800X3D’s for hosting top-performance Minecraft servers.
How do the rear OCP slots work shared with the nodes?
I am wondering both physically and logically? If each one of those needs 4x 4X PCI-E cables for full bandwidth, how do these get connected?
@George these will require OCP NIC from the 3.0 standard with multi host capabilities like a Connect-X 6, you get one or two physical connections into the NIC and then the NIC has vSwitch/vRouter capabilities that present a separate NIC to the intrrnal hosts, typically 4 per card (that’s why you need 3 for 10 nodes)
Notably, each node seems to only have a Gen4 x4 connection to the multi-host NICs, and it’s from the B650 chipset, not direct from CPU. That means you’re capped at 60 Gbps, and it’s shared with all the other chipset functions. I also wonder if it will impact things like SR-IOV & VFIO (because of iommu groups).
Ten individual separate PCs in one box. Each PC can be leased (rented?) to a user who logs in to an online cloud hosting service and remotes into their own online cloud PC. Ten users at once, each to their own respective PC. The one box requires less power to operate than ten separate boxes, each with its own PSU and corresponding PDU socket to do the same job. Makes sense to me. I like it!
Also, it’s a physical hardware PC, so fewer issues with setting up virtualization (although that’s possible here too). Very neat server design. I can see this being useful for businesses running their own federated cloud PC services to users who need to have a separate work computer but maybe can’t afford to buy a whole new system on their dime.