- Sparkle’s close PCI 5.0-Design offers lower costs to AI and data research
- Up to 32 Intel GPUs deliver 768 GB VRAM to scalable AI -computing
- 10.800W Overflowing power system and advanced cooling enables continuous heavy workload
Taiwanese graphics card manufacturer Sparkle has taken the wrapping of its new close, scalable GPU server that can offer an alternative to expensive AI accelerator systems such as NVIDIA’s H100, H200 and B200 Data Center GPUs.
The C741-6U-double 16P is designed to house up to 16 ARC Pro B60 double graphics card, giving a total of 32 Intel GPUs and as much as 768 GB combined VRAM.
Each ARC Pro B60 double has two battlemage BMG-G21 GPUs, and when fully equipped, the system delivers a total of 81,920 GPU cores.
Advanced cooling
The server supports both 24 GB and 48 GB GPO models, paired with 4th or 5th generation Intel Xeon scalable processors.
It uses the Intels C741 chipset and supports up to 32 DDR5 memory sites, allowing 256 GB per day. Module for intensive workload.
Sparkle has built the system around a 6U chassis with strong power and cooling capacity.
The top configuration uses five 2,700W titanium current supplies and reaches a combined 10,800W output, while the smaller version includes four 2,400W devices rated at 7.200W.
With as many as 15 fans installed, the system has clearly been built to control heat effectively during continuous workloads.
Each GPU is connected to the CPU through a PCIe 5.0 x8 interface, either directly or via a microchip -switch that maximizes the data taiser while avoiding potential bottlenecks caused by shared links.
There is support for double M.2 slots and several NVME and SATA buges for local storage.
Although there is no way to know at this point in how the ARC Pro B60 will work in large -scale inferencies or training tasks, the flexibility offered by hardware could appeal to researchers and developers seeking affordable parallel computation power.
Not surprisingly, Sparkle has not revealed pricing for the C741-6U double 16P, but potential buyers can conduct a purchase survey on the site.
Via Videocardz
Follow Techradar on Google News and Add us as a preferred source To get our expert news, reviews and meaning in your feeds. Be sure to click the Follow button!
And of course you can too Follow Techradar at Tiktok For news, reviews, unboxings in video form and get regular updates from us at WhatsApp also.
