You are here

Scheduling & Job Management: How to Get the Most from a Cluster (3-part series)

Date:

Wednesday, November 25, 2015 - 8:00 MST to Friday, November 27, 2015 - 8:00 MST

Speaker Info:

Kamil Marcinkowski
WestGrid Site Lead / System Administrator
University of Alberta

Description

This online workshop focuses on getting the most out of a cluster using a scheduler. Split into three parts, each two-hour session will provide hands-on training and offer users a chance to experiment with job submission techniques. Held over three days, each of the three sessions will be delivered online as a two-hour webinar.

Wednesday, November 25 - Friday, November 27
1:30 - 3:30 pm MST each day
Participants will connect to the webinar using Vidyo
(connection instructions will be sent to all registrants)

Part 1 will discuss how a scheduler works and how to use the knowledge to your advantage. Participants will practice how to submit basic jobs, with a focus on productivity. Other topics covered include: 

  • optimizing walltime 
  • MPI and OpenMP jobs 
  • job arrays 
  • Interactive Jobs

Part 2 will discuss how to submit and run more advanced jobs. Participants will practice examining jobs, and understand their state. Knowledge and practice of the topics covered in Part 1 is a prerequisite for attending this workshop. Other topics covered include:

  • Jobs and memory 
  • partitions 
  • gpus 
  • software licenses, job dependencies, accounting groups and advanced requests

Part 3 will discuss how a cluster chooses which jobs are to be run first by examining the topics of fairness, priority, and reservations. Participants will examine the state of the cluster and their jobs in order to troubleshoot problems. Knowledge and practice of the topics covered in Parts 1 and 2 is a prerequisite for attending this workshop.

Course Materials:

Who Should Attend:
This session is intended for current WestGrid account holders who have had some experience submitting jobs but would like to learn how to increase their job productivity and efficiency. Familiarity with logging in to WestGrid systems, using UNIX and basic job script commands is recommended.

Participants should:

  • Have a WestGrid account and have experience logging into the Jasper cluster
  • Be somewhat proficient in U/Linux, i.e.:
    • Know what a man page is.
    • Can edit, copy, and delete files.
    • Can use top and ps to see resources used of a process.
    • Know what unix environment variables are, how to set and display them.