Our Computers has extensive experience in building server systems of various classes. The company is currently focusing on high-performance rack systems in 1U, 2U and 4U form factors based on Intel Xeon and AMD EPYC processors. They can be divided into 3 groups:

  • General purpose servers
  • Multi-site servers
  • GPU Servers

General purpose servers are intended for use as file, cloud, virtual, computing servers, database servers. They feature state-of-the-art Xeon and EPYC server processors, significant RAM support, and hot-swappable 2.5 ”disk drives. The latter can have both a SATA interface and the fastest NVMe interface for solid state drives. There are also two connectors for connecting SATA-DOM modules. There is a significant set of network interfaces - 10 GbE Ethernet, SFP +, IPMI 2.0, OCP 3.0.

Multi-site servers are a combination of 4 servers in a single 2U chassis. This is useful when you need to save rack space. Shared power supplies (2 2600 W power supplies installed) also save space and energy for powering and cooling servers. This layout is especially useful for compute or failover clusters, or for virtualization. The most popular solutions are from Microsoft, Stratus, Citrix, VMware. All of them imply redundant hardware resources, so such layouts of Polywell Computers servers are especially beneficial in such applications Each of the four nodes supports the latest Intel or AMD processors, up to 4 TB of memory, 6 disk drives 2.5 ”, 1 slot OCP 3.0 for network modules , one IPMI 2.0 (Intelligent Platform Management Interface) module.

GPU servers are designed for high performance computing (HPC), GPU compute clusters, artificial intelligence, and deep machine learning. Available in 2U form factor with 4 GPUs Nvidia® HGX A100 GPU 40GB / 80GB, NVLink, NVSwitch and 4U form factor with 8 GPUs Nvidia® HGX A100 GPU 40GB / 80GB, NVLink, NVSwitch. Also, depending on the form factor, they can support up to 8 TB (12 TB) of memory, have up to 4 (6) drives in the 2.5 ”form factor, 2 x 10GbE network adapters or 1 OCP 3.0 slot for network modules. Powered by GPU-powered Polywell Computers servers, AI acceleration is up to 20x and HPC acceleration up to 2,5x over previous generations of GPUs.