HBS Compute Grid 2.5 is arriving Monday 11/19!

November 9, 2018

Greetings! We are happy to announce the opening of HBS Compute Grid 2.5, beginning Monday, Nov 19th.

Based on both research computing trends and recommendations from the HBS research computing environment assessment, Research Computing Services (RCS) has been working closely with HBS IT to make improvements to our local compute grid. This new compute grid, v2.5, provides the following updates and enhancements:

  • Improved compute capacity through more hardware and better use of existing hardware
  • Significantly fewer restrictions on compute capacity
  • Increased safeguards to prevent CPU spillover and memory problems
  • Newer OS and software versions, and improved usability
  • More software titles, including GitKraken, a GUI Git version control application, and Spyder, a Python IDE
  • Better command-line submission scripts, to improve productivity.

With the successful testing over the past few months, we will open the new environment for general use, and will shortly begin to decommission the old environment. We plan to transition everyone over to the new environment by the December holiday.

How will this affect my work?

We hope in a positive manner! With more compute cores, and fewer restrictions, we hope you can get more research done. We've also put in place safeguards to isolate your work (jobs) from other jobs that might be misbehaving. But we've tried to engineer the new environment to be at least as user friendly as the current environment. We are also providing several helpful guides so you can understand the changes. Finally, this upgrade affects only the compute -- home folders and projects spaces are unchanged.

Beginning 11/19, we will add more processing capabilities in the new environment by removing compute nodes in pairs from the current environment on a regular basis. Barring any unforeseen problems, we wish to have everyone using the new environment before we break for the holidays.

What will I need to do differently?

The login names for accessing the new environment have changed, and are documented in the Grid 2.5 Welcome instructions (https://grid.rcs.hbs.org/compute-grid-25). Additionally, if you are using NoMachine, the interface has changed slightly. Take a few moments to review these changes and the other documents on our Welcome page. If you are mapping network drives / mounting network volumes to your local computer, this will not change.

Are there any disadvantages?

In the old environment, interactive sessions and background jobs could run without time limits. In order to accommodate the reduced restrictions on compute and to be in line with best practices, we've limited the amount of time that applications and scripts can run. This is documented in full detail in our Major Usability Changes webpage (https://grid.rcs.hbs.org/major-usability-changes).

What if I have a problem?

We’re hoping that you won’t have any! We’ve had other researchers and our own RCS team using the new environment since late August. Testing has been going smoothly, and we believe that we have caught most of the problems. But in complex systems, one cannot catch every problem in advance. As is the custom now, please contact RCS if you are having difficulties. Additionally, please try to describe your problem in detail; error messages and screen captures of the problem are immensely helpful!

What if I want to keep using the current environment?

Unfortunately, that is not possible. As mentioned previously we will slowly migrate compute to the new environment. At the end of the transition period, there will be no compute nodes in the current environment left to use.

Sounds great! How can I get started?

After reading our Welcome document, no need to wait until 11/19. Just dive right in!

Please contact RCS if you should have any comments or questions, and on behalf of HBS IT and the entire RCS staff, we thank you for your patience during this exciting transition period.

Best,

The RCS Staff

See also: News