[ad_1]
Right now, we introduced two new GPU servers, the ProMagix G480i and G480a. Designed particularly for compute intensive duties like simulation, rendering, analytics, deep studying, and AI, these rackmount supercomputers take our efficiency PC heritage and crank it as much as 11. Let’s discover.
What’s GPU Computing?
With the introduction of CUDA cores, NVIDIA modified the sport. This allowed for the computation of advanced calculations by breaking them up into smaller ones and working them in parallel on the various CUDA cores of a GPU or a number of GPUs, reasonably than the CPU. By spreading out these purposes over hundreds of cores, sure calculations might be sped up exponentially. Now with the discharge of Tensor and RT cores alongside CUDA, NVIDIA GPUs are capable of relegate particular render or AI computation to cores optimized for these duties, additional accelerating them.
ProMagix G480i GPU Server
First is the G480i, a customized HPC/AI server designed for versatility and scalability, powered by twin 4th Gen Intel Scalable Xeon processors, making it superb for purposes accelerated by the Intel instruction set. In-built a 4U rackmount kind issue, the G480i helps as much as 8 x PCIe Gen 5 GPUs. And like every little thing we do, the G480i is assembled and supported by our group of US-based specialists.
Base Specs:
Helps as much as 8 x Twin slot Gen5 GPU playing cards
2 x 4th Gen Intel® Xeon® Scalable Processors
12-Channel RDIMM DDR5, 24 x DIMMs
2 x 10Gb/s BASE-T LAN ports
2 x M.2 slots with PCIe Gen3 x4 and x1 interface
8 x FHFL PCIe Gen5 x16 slots for GPUs
2 x FHFL PCIe Gen5 x16 slots
3+1 3000W (240V) 80 PLUS Titanium redundant energy provides
ProMagix G480a GPU Server
Subsequent, the AMD Epyc based mostly G480a. This 4U server is designed to additionally maximize multithreaded CPU efficiency alongside GPU by using twin Epyc 9004 sequence processors with as much as 384 whole processing threads. Prefer it’s Intel counterpart, the G480a additionally helps as much as 8 x PCIe Gen 5 GPUs with hand meeting right here in Richmond.
Base Specs:
Helps as much as 8 x Twin slot Gen5 GPU playing cards
2 x AMD EPYC™ 9004 sequence processors
12-Channel RDIMM DDR5, 24 x DIMMs
Twin ROM Structure
2 x 10Gb/s BASE-T LAN ports (Intel® X710-AT2)
2 x M.2 slots with PCIe Gen3 x4 and x1 interface
8 x FHFL PCIe Gen5 x16 slots for GPUs
2 x FHFL PCIe Gen5 x16 slots
3+1 3000W (240V) 80 PLUS Titanium redundant energy provides
Suppose one among these new methods might assist speed up your workflow? We’d love to speak to you about it. Click on to study extra concerning the Xeon powered ProMagix G480i or AMD powered ProMagix G480a. Or, name our skilled gross sales group at 804-419-0900 for session and help with configuration questions.
The next two tabs change content material beneath.
Josh has been with Velocity Micro since 2007 in varied Advertising, PR, and Gross sales associated roles. Because the Director of Gross sales & Advertising, he’s accountable for all Direct and Retail gross sales in addition to Advertising actions. He enjoys Seinfeld reruns, the Atlanta Braves, and Beatles songs written by John, Paul, or George. Sorry, Ringo.
[ad_2]