Hardware

Gigabyte launches two nvidia tesla-based 4u gpu servers

Table of contents:

Anonim

GIGABYTE has announced a couple of new servers focused on hosting a large number of GPUs, with up to 8 GPUs for the G481-S80 and up to 10 GPUs. on the G481-HA0, offering data centers one of the highest GPU densities in this format available on the market.

GIGABYTE Launches Two Servers Focused on Deep Learning

The new servers feature dual Intel Xeon Scalable processors with up to 28 core CPUs and the highest 205W TDP CPU requirement for each socket. With deep learning-based systems, data centers need high-density machines to meet the needs of their customers, as well as to save space.

The GIGABYTE G481-S80 can accommodate up to eight SXM2 form factor GPUs from NVIDIA's Volta V100 or Pascal P100 lines, with support for NVLink interconnect, providing increased bandwidth, more links, and improved scalability. A single NVIDIA Tesla V100 GPU can support up to six NVLink links with a speed of 25GB / s in each direction and a total bandwidth of 300GB / s. Meanwhile, the G481-HA0 is capable of supporting a total of 10 PCIe GPU cards.

For storage, the G481-S80 has 10 2.5-inch hot-swappable bays in the front, capable of supporting four NVMe drives through a U.2 connection and six drives through SATA / SAS. The HA0 goes further and offers 10 2.5-inch hot-swap drive bays capable of accommodating up to eight 2.5-inch NVMe drives and two 2.5-inch SATA / SAS drives. Beneath the 2.5-inch storage drive area are 12 additional bays for 3.5-inch hot-swap SATA / SAS drives.

Neither prices nor availability were mentioned; however, their product pages are on the website, so we hope to see you soon.

ANANDTECH Font

Hardware

Editor's choice

Back to top button