AWS and OpenAI ink a $38B deal to provide NVIDIA GB200 and GB300 compute further showing the demand for AI compute and power
Meta shared its vision for GPU scale compute with 1PB SSDs and a new E2 form factor to maximize QLC NAND SSD capacities
We take a look at 25G SFP28 optics and discuss LR versus SR optics and why that distinction matters if you want to connect devices
We review the Gigabyte B343-C40-AAJ1, a 3U 10-node AMD EPYC 4005 and Ryzen 9000 server that seeks to drive up density and drive out costs
We take a look at the GL.iNet Comet GL-RM1, a little box that lets you easily remotely manage systems as if you were sitting next to them
At OCP 2025, we saw the liquid-cooled Meta IcePack rack for next-generation AI and high-performance networking applications
In our ASRock Rack 8U8X-GNR2 SYN B200 review, we see how this NVIDIA HGX B200 8-GPU server exemplifies ASRock Rack's engineering flair
The upcoming 2026 NVIDIA BlueField-4 DPU will sport 64 cores along with 800G networking for Vera Rubin AI clusters
ORNL announced the AMD-powered Discovery and Lux systems alongside HPE's announcement of its new supercomputing portfolio
Qualcomm announced new integrated AI racks, with the Qualcomm AI200 sporting up to 768GB LPDDR cards, and a 200MW AI deployment