Introduction

Instruction for the Cluster

Hello, this is the information page about the cluster in the Institute of Nuclear and Particle Physics (INPAC), school of physics and astronomy of SJTU.

If you have an account, you can login to the server easily with:

$ ssh -Y your_user_name@bl-0.inpac.sjtu.edu.cn

If you don’t have an account, you can apply one by sending an email to the administor. Please ask related people (your colleague, your advisor …) for the email address.

Currently, we have two clusters sharing the same set of user names. That means you can use both of them. You can only reach the login node “bl-0” from outside. If you want to use the second one, please try to connect to it with ssh -Y bl-2-1 after you get into bl-0. And “bl-2-1” is the login node of the second cluster.

Password

To change password, first use your current password to get a Kerberos ticket-granting ticket:

$ kinit your_user_name

Then:

$ ipa user-mod your_user_name --password

Enter the new password.

The password can only be changed once in an hour, the minimal length is 8, and cannot re-use the previous three passwords.

Guide for Job Submission

HTCondor is used for computional job management.

HTCondor is a specialized workload management system for compute-intensive jobs. Like other full-featured batch systems, HTCondor provides a job queueing mechanism, scheduling policy, priority scheme, resource monitoring, and resource management. Users submit their serial or parallel jobs to HTCondor, HTCondor places them into a queue, chooses when and where to run the jobs based upon a policy, carefully monitors their progress, and ultimately informs the user upon completion.

Users should NOT execute heavy load programs (For example, a program uses one CPU 100% for over 5 minutes.) on the interaction server directly. Please submit your job to the HTCondor system.

What to know how to submit your job? Please read the Condor Quick Guide.

Softwares

ROOT

To use the pre-built ROOT6 from CERN:

source /sw/root/6.22.02-gcc4.8/bin/thisroot.sh

You can find more versions of ROOT under /cvmfs/sft.cern.ch/lcg/app/releases/ROOT.

GEANT4

Currently, two version of Geant4 are installed on the cluster. They are geant4.9.6p02 and geant4.9.4p04.

For more information, please see the page of Geant4 Guide

Hardware

Overview of cluster processors.

Cluster

Nodes

CPU

#cores

Memory

InfiniBand

bl-3

16

Xeon E5-2650v3

20

128GB

N/A

bl-4

16

Xeon Gold 5220

36

192GB

Mellanox 56Gb

bl-hd-1

1

Xeon Gold 6242

64

768GB

Mellanox 100Gb

bl-hd-2

1

Xeon Gold 6148

80

768GB

Mellanox 100Gb

Overview of GPU Hardware.