GPU Server

GPU Virtual Servers Optimized for Cloud Computing

GPU Server allows users to assign only the amount of necessary virtual computing resources at the time when needed, without having to purchase CPU, GPU, memory or other infrastructure resources from servers. It is ideal for workloads that require fast computation such as AI model experiment, prediction and inference, and users can select resources with optimal performance depending on job type and size.

Overview

01

04

Service Architecture

    General Computing
  • Application
  • Operating System
    Virtual Computing(GPU Server)
  • Applications
  • Guest OS
  • Virtualization (Hypervisor)
    • GPU / NVSwitch / NVLink
    • CPU, Memory, Network I/F, Storage

General Computing → Virtual Computing

Key Features

  • VM creation/management

    - Select GPU for work type and required spec (NVIDIA A100 or V100 GPU)
    - Offer various OS types and standard images for creating resources conveniently (e.g. Windows, RHEL and Ubuntu)
    - Securely connect to the OS using key pairs
    - Manage tag creation/revision

  • Storage and network connection

    - Provide additional storage besides OS disk
    - Set up subnet/IP and enable the use/turn off NAT IP setting
    - Provide Security Group integration setting

    • Billing
    • Without contract : Hourly billing based on the time allocated after requesting for resources
    • With contract : Discount applied according to 1/3 year contract terms. Fixed monthly charges for the amount of requested resources
      * Penalty is charged for early termination within the contract period
    • Basic offering and add-on options
    • Basic offering : OS S/W, 100GB for OS installation disk, and 1 IP
    • Additional charges for add-on offering/options, including reserved IP
Let’s talk

Whether you’re looking for a specific business solution or just need some questions answered, we’re here to help

Share