SIGN-IN

Press Releases

Orca has been converted to the same software environment as Graham, and now requires your Compute Canada username and password. Software access and job submission works the same way as it does on Graham.

You can connect to the new Orca via orca.computecanada.ca

The legacy SHARCNET software stack is still available and can be used by issuing the following two commands:
module purge --force
export MODULEPATH=/opt/sharcnet/modules

Please report any software issues you find by email to help@sharcnet.ca

On the new Orca you will have the same /home and /project data as you have on Graham. Because user ID numbers are different between SHARCNET and Compute Canada we must convert file ownerships before making /scratch available. All files will eventually be transferred for you and no scratch expiry will happen this month.

Your old /home and /work are still available on the old Orca via orca.sharcnet.ca, dtn.sharcnet.ca and other SHARCNET clusters.



Starting Tuesday, August 21, 2018 at 7 a.m. ET, the power feed to SHARCNET’s main data centre is being upgraded. This will mostly impact Graham but it could also affect storage and general operations of other SHARCNET clusters. We will be posting status messages on our website.

After the outage, on Aug 29th, we will be bringing the orca cluster back into operations with a new software stack. The cluster will be renamed to orca.computecanada.ca and it will have the same scheduler, software availability and storage as Graham. It will maintain unique /scratch storage but /home and /project will be the same storage as Graham. We will maintain a very small orca.sharcnet.ca for a few weeks for any users that need to move files.

Please email help@sharcnet.ca if you have any questions.



SHARCNET is pleased to announce the results of its Round XII Dedicated Programming Support competition. The primary objectives of this program are to enable key research projects with the potential for exceptional and lasting impact that require significant programming support to proceed, and facilitate optimal exploitation of SHARCNET’s or Compute Canada’s computing infrastructure for internationally leading research.

In this Round, programmer allocations have been made to the following researchers:

  • Hugh Couchman, Physics and Astronomy, McMaster University
  • Ranil Sonnadara, Surgery, McMaster University

SHARCNET plans to run another competition in the Fall of 2018. Please visit our DP website for more information. Congratulations to our Round XII awardees!



SHARCNET is issuing a call for proposals for Dedicated Programming Support Round XII. For this round, applications are encouraged that satisfy the programme objectives and priority will be given to proposals that meet one or more of the following conditions:

  • Propose an innovative project that will leverage the capabilities of the GP3 system, “Graham”, especially GPUs and FPGAs.
  • Propose a programme of work that deals with the efficient processing of large, heterogeneous datasets (“big data”) using a variety of data mining, machine learning or other analytics software.

Applications are submitted via SHARCNET’s webportal and are due by June 22, 2018 at 11:59 pm EST. Consultation with a SHARCNET HPTC prior to submission is required. Note that users must have a SHARCNET webportal account in order to access the online form.

Due to the current workload of SHARCNET programming staff, only a limited number of proposals are expected to be awarded in this round. For additional information, please refer to the application guidelines. Questions should be addressed to research-support@sharcnet.ca.



We are pleased to announce 2018’s Compute Ontario Summer School on High Performance Computing. The SHARCNET offering (West) is being held May 28 – June 1, 2018 at Western University.

Register online NOW. Online registration is open until 23:59, Fri, May 25, 2018.

A long-running event for SHARCNET, Summer School was introduced to users over a decade ago. Summer School is now an Ontario-wide event taking place at three campuses every year. Summer School Central will run June 11 – June 15, 2018 at the University of Toronto, and Queen’s University will host Summer School East from July 30 – August 3, 2018. Note that the courses offered by the three sites may be slightly different. Please check the site specific programs for details.

While there is no fee to attend Summer School, registration is required. For more information, check out our Summer School preview.



With just one month to go, registration is gaining momentum for Ontario’s Advanced Research Computing Congress! Building on the 30-year history of High Performance Computing Symposium (HPCS) – Canada’s premiere HPC Conference – Compute Ontario has launched Ontario Advanced Research Computing Congress (OARCC). This reinvigorated event offers researchers, policy makers, technical, and administrative leaders a forum to showcase the latest research trends, discoveries, and technical achievements in the ARC sector. OARCC is a multidisciplinary gathering of thought leaders and practitioners in high performance computing. Register today.



SHARCNET is pleased to announce the results of its Round XI Dedicated Programming Support competition. The primary objectives of this program are to enable key research projects with the potential for exceptional and lasting impact that require significant programming support to proceed, and facilitate optimal exploitation of SHARCNET’s or Compute Canada’s computing infrastructure for internationally leading research.

In this Round, programmer allocations have been made to the following researchers:

  • Susan Brown, English and Theatre Studies, University of Guelph
  • Natalia Ivanova, Physics, University of Alberta
  • Marek Stastna, Applied Mathematics, University of Waterloo

SHARCNET plans to run another competition in late Spring of 2018. Please visit our DP website for more information. Congratulations to our Round XI awardees!



On Jan 30 all legacy SHARCNET systems (this does not include “Graham”) will be unavailable for approximately 2 days. During this time we will move users off of some older, unstable storage to a new storage array. All files will be moved and users should not notice any difference when logging back in after the outage. We will also be patching the systems to address the recently identified Spectre and Meltdown security vulnerabilities.

Our team constantly patches issues and this is mostly done transparently to users. In this case a full reboot is required and all jobs will be lost.

This issue, along with the majority of recent security alerts are called local exploits. That means the systems can not be compromised by a remote attack, it must be done after correctly logging into the cluster. All users are encouraged to protect their passwords and ssh keys. Security is everyone’s responsibility.

How to choose a password.