Environment Requirement

Hardware Environment

Before the installation, check the hardware configuration listed in Table 1.

Table 1 Hardware environment

Type

Configuration Reference

Server (single-node)

Atlas 800 training server (model 9000)

Server (cluster)

Compute node: Atlas 800 training server (model 9000)

Storage node: storage server

Memory

  • Recommended memory size: ≥ 64 GB
  • Minimum memory size: ≥ 32 GB

Drive space

≥ 1 TB

For details about the drive space plan, see Table 3.

Network

  • Out-of-band management (BMC): ≥ 1 Gbit/s
  • In-band management (SSH): ≥ 1 Gbit/s
  • Service plane: ≥ 10 Gbit/s
  • Storage plane: ≥ 25 Gbit/s
  • Parameter plane: 100 Gbit/s

Software Environment

Before the installation, install the software listed in Table 2.

Table 2 Software environment

Software

Version

Installation Position

How to Obtain

Operating system (OS)

  • CentOS 7.6 Arm
  • CentOS 7.6 x86
  • openEuler 20.03 Arm
  • openEuler 20.03 x86
  • openEuler 22.03 Arm
  • openEuler 22.03 x86
  • Ubuntu 20.04 Arm
  • Ubuntu 20.04 x86
  • Ubuntu 18.04.5 Arm
  • Ubuntu 18.04.5 x86
  • Ubuntu 18.04.1 Arm
  • Ubuntu 18.04.1 x86
  • Kylin V10 SP2 Arm
  • Kylin V10 SP2 x86
  • UOS20 1020e Arm

All nodes

-

Python

≥ 3.7

Compute node

Installed by users

Torch

2.7.1

Compute node

Installed by users

MindSpore

≥ 2.7.0

Compute node

Installed by users

OS Drive Partitions

Table 3 lists the recommended OS drive partitions.

Table 3 Drive partitions

Partition

Description

Size

Bootable Flag

/boot

Boot partition

500 MB

on

/var

Partition for storing data generated during software running, such as logs and cache

> 300 GB

off

/

Primary partition

> 300 GB

off