|Table of Contents|
Consulting is available free of cost to everyone and we welcome you to reach out! Our services include, but are not limited to:
- Help for new users getting started with our resources. We know using HPC systems for the first time can be intimidating so scheduling an in-person meeting where you can ask loads of questions can help a lot
- General issues that may occur, e.g., why is my job spending so long in the queue, where can I find information on topic <A>, why is this strange and unexpected thing happening, etc.
- Advice on debugging software, package installations, and job failures
- Advice on code optimization and utilizing our resources more effectively/efficiently
How to Reach Us
We use ServiceNow and can be reached with a support ticket.
When We're Available
Our consultants are available during standard working hours (9:00 am-5:00 pm, Monday-Friday) and will respond to any user requests as promptly as we are able. Note: our consulting staff is generally unavailable on weekends, staff holidays, and after hours.
When Can I Ask for Help?
Any time! But first, we encourage you to:
Look through our online documentation - There's lots of information to help get you started that may answer your question or help give you a better idea of what to ask.
How Can I Effectively Write a Support Request?
Glad you asked, helping us help you goes a long way and can give you better answers faster. Some general rules of thumb:
Detail detail detail - A full error long may seem like a lot to send, but the more information we have, the more likely we are to be able to diagnose and/or replicate your issue. Information that can help us includes:
- The cluster you're using
- The modules you have loaded
- The exact commands you're running
- Full error logs
- How you're connecting to the system (is it through a terminal, Open OnDemand shell, or virtual desktop?)
- Your batch submission script. This can be included either as a text file or by providing its full path and filename on the system.
Use reply-all to email chains - We will cc hpc-consult in our responses so that our (small) consulting team is able to view the issue and contribute.
Provide context - There's a common support issue called The XY Problem. Say you have a problem and try to solve it yourself but the attempted solution produces an additional problem. Submitting a ticket requesting help with the attempted solution without information about the original issue can lead to more confusion. Knowing the full context will help us be able to more effectively assist you.
Submit a ticket for your questions - If you reply to general system announcements or send emails to an HPC staff member's private inbox without cc'ing hpc-consult, your ticket may get lost in the shuffle and go unanswered (especially if your ticket goes to a staff member who has retired or is on vacation). Submitting a ticket will ensure we have a record of your question and will get to it as promptly as we are able.
We host virtual, drop-in office hours every Wednesday from 2pm to 4pm. Drop by to visit with our consultants to ask any questions you have about using HPC resources. It might be a bunch of getting started questions, or you might want to share your screen to walk us through a particular problem you're hung up on. We have private spaces for one-on-one consults. You can join us in Gather Town at this link.
If you have never used Gather Town before and would like additional information, check out this page.
|HPC Quick Start Guide||A walkthrough guide that demonstrates how to access the system, load software, and write and submit jobs batch jobs|
|Intro to HPC Video Presentation||A video recording of our Intro to HPC workshop designed to provide an overview of the system and walk users through submitting their first batch job|
|In-Person Workshops||Throughout each semester, we provide workshops that users can register for. Topics include: Intro to HPC, Intro to Machine Learning|
|Script Examples||Our GitHub site that has a number of script examples for different kinds of workflows that can be executed on HPC.|