WML CE prerequisites

IBM Watson® Machine Learning Community Edition is optimized to leverage the unique capabilities of IBM® Power® Systems accelerated servers.

System support

WML CE is supported on the following systems:

  • IBM Power System IC922 with or without NVIDIA Tesla T4 GPUs
  • IBM Power System AC922 with or without NVIDIA Tesla V100 GPUs
  • IBM Power System S822LC with or without NVIDIA Tesla P100 GPUs
  • x86_64 systems with or without NVIDIA Tesla V100, P100, or T4 GPUs
Table 1. Differences in support on supported systems
  IBM Power x86_64
SnapML Up to two GPUs and single node Up to two GPUs and single node

Required third party components

WML CE requires some additional third-party software components.

Table 2. WML CE additional software
Component Version
Red Hat Enterprise Linux
  • RHEL for POWER9™, version 7.6 (also referred to as RHEL-alt)
  • 7.7
Ubuntu 18.04.1
NVIDIA GPU driver

(if GPUs are present)

440
Anaconda 2019.10
Note:

To address security concerns with data exposed by the GPU performance counters, NVIDIA GPU driver versions 418.43 and later restrict profiling GPU performance to administrators. For more information about this issue and for instructions to allow other user ids to do profiling. See the NVIDIA documentation.

Additional information

Developer resources can be found at IBM Cognitive Systems developer portal.

Have questions? Try asking in IBM Cognitive Systems space on dW Answers.