Skip to main content

CTE-POWER

System Overview

CTE-POWER is a cluster based on IBM Power9 processors, with a Linux Operating System and an Infiniband interconnection network.

It has the following configuration:

  • 2 login node and 52 compute nodes, each of them:
    • 2 x IBM Power9 8335-GTH @ 2.4GHz (3.0GHz on turbo, 20 cores and 4 threads/core, total 160 threads per node)
    • 512GB of main memory distributed in 16 dimms x 32GB @ 2666MHz
    • 2 x SSD 1.9TB as local storage
    • 2 x 3.2TB NVME
    • 4 x GPU NVIDIA V100 (Volta) with 16GB HBM2.
    • Single Port Mellanox EDR
    • GPFS via one fiber link 10 GBit

The operating system is Red Hat Enterprise Linux Server 7.5 alternative.

Other relevant information

caution

The most recent installed version of CUDA is the 10.2. Installing newer versions would require a cluster-wide update of the operating system to RHEL 8, which is something not planned to do as of now.

  • CUDA version: 10.2
  • CUDA driver: 440.33.01
  • Mellanox IB 4.3-4
  • IBM Firmware: 920.10

The interconnection between CPU and GPU is done using NVLink 2.0.

Compiling applications

For compiling applications the system provides GCC version 4.8.5, IBM XL compilers for C/C++ v13.1.6 and for Fortran v15.1.6.

Via modules you can find other compilers such as:

  • GCC/7.3.0
  • PGI/18.1

Connecting to CTE-POWER

The first thing you should know is your username and password. Once you have a login and its associated password you can get into the cluster through the following login nodes:

  • plogin1.bsc.es
  • plogin2.bsc.es

This will provide you with a shell in the login node. There you can compile and prepare your applications.

You must use Secure Shell (ssh) tools to login into or transfer files into the cluster. We do not accept incoming connections from protocols like telnet, ftp, rlogin, rcp, or rsh commands. Once you have logged into the cluster you cannot make outgoing connections for security reasons.