ZOTAC's new HPC server can take 2 x Intel Xeon CPUs, 10 x GPUs and 12,000W of power

ZOTAC has just announced its expanded GPU Server Series systems. The first of the series is the Enterprise lineup, which offers companies affordable, high-performance computing solutions for countless applications, including AI.

5

These new ZOTAC systems are aimed at core-to-core inferencing, data visualization, model training, HPC modeling, simulation, and AI workloads. ZOTAC's new family of GPU servers comes in varying form factors and configurations, with the Tower Workstation and Rack Mount Servers both offered in either AMD EPYC or Intel Xeon CPU form.

There's support for a huge 10 x GPUs, with a modular design that makes it easier to get in and configure the system, with a high space-to-performance ratio, industry-standard features like redundant power supplies and various cooling options, ZOTAC has your back with its new GPU Server Series.

ZOTAC explains: "Selections of the lineup include the ZRS-3220M4 4U Intel Rack-Mounted GPU Server, an affordable HPC server option that can accommodate up to 10 GPUs with its 10 PCIe x16 slots, two 5th and 4th Gen. Intel Xeon Scalable Processors for intense workloads, and 4+1 2000W (240V) 80+ Platinum redundant power supplies to ensure minimal downtime and reliable performance".

ZOTAC explains its new Essential Series: "optimized for cost-effectiveness and flexibility offered in BOLT Tower Workstation and 4U rack server form factors. Ideal for system integrators and businesses seeking a good price/performance ratio with Intel 3rd and 4th Gen Processor support for Generative AI inference, data science, 3D content creation, and game development applications".

5

Advanced Series: "represented by a wider selection of 4U rack servers, the series is a step up from the Essential Series, and leverages PCIe Gen 5 slots for increased throughput, giving the series the high performance to take on more demanding tasks and extended operational capabilities".

5

Premier Series: "available in both 4U and 8U Rack Mount servers, the Premier Series is tailored for the most challenging workloads such as large language model (LLM) and Cloud AI model training, with the ability to leverage up to eight Server PCI Express Module (SXM) or OCP Accelerator Module (OAM) GPUs from all major vendors. The Premier series offers top-tier performance and reliability, as well as up to two 10Gb/s LAN Ports and increased storage capacity among other upgrades from the Advanced Series".