SSCC News December 2023

Introducing Open OnDemand

We are happy to announce a pilot of Open OnDemand, a new way to access interactive Linux servers. OnDemand lets you run all the same programs as Linstat and submit jobs to Slurm, but with a graphical “desktop” interface similar to Winstat or a regular PC or Mac. You can also start a terminal program and then use the command line just like you’re used to. To access OnDemand all you need is a browser–no need for special software like X-Win32 or XQuartz. Just get on the SSCC network using VPN (or log into Winstat), go to ondemand.ssc.wisc.edu and log in using your SSCC credentials. The SSCC Knowledge base can walk you through how to log in to OnDemand and how to carry out common tasks using the OnDemand interface.

While this is a pilot, if all goes well we anticipate Open OnDemand becoming the primary way to use interactive Linux servers at the SSCC.

Spring Training

The SSCC’s spring training schedule is now available. The core Data Wrangling in R, Stata, or Python workshops will be taught in mid-January, between semesters, and online so you can take them even if you’re not back in Madison yet.

Workshops are free, but registration is required. See the training schedule for details and to register.

Speeding up Winstat Logins (But Resetting Settings)

During the December 27th monthly downtime, we will be making an update to our Winstat and WinSilo servers that will result in a significant decrease in login times! Unfortunately, it is not possible to carry over settings to the new system. This will impact things like browser histories, bookmarks, and any other program customizations stored in your Winstat profile. We strongly recommend using Firefox or Microsoft Edge to sync your profile to make sure your bookmarks and profiles get backed up prior to the update.

 Winter Tech Update

On Saturday, January 6, from 9:00AM to 5:00PM, SSCC staff will take Winstat, Linstat, Silo, and Slurm offline for updates. SSCC disk space will also be unavailable (U:, V:, X:, Z:, /project, etc.). Any running jobs will be terminated, though Slurm will restart some jobs automatically once the downtime is over. During this time all SSCC software will be updated.

R users: if you have installed additional packages, you will need to reinstall the latest versions after the upgrade.

Python users: we will install the latest Anaconda version, including all the packages it contains. If you use a lot of additional packages, or sensitive packages like pytorch, we strongly recommend you use tools like conda or pyenv/venv to create a stable environment for each of your projects.

Starting on January 6th, multifactor authentication (MFA) will be required to access Linstat unless you are on Winstat or using the SSCC VPN (which already requires MFA).

MFA will also be required to connect to ftp.ssc.wisc.edu (whether directly or using SecureFX or other clients). If you are on the SSCC VPN you can use FTP without MFA by making an SFTP connection to linstat.ssc.wisc.edu.

This is part of a gradual process of requiring MFA to access all SSCC resources. MFA is a critical security measure in today’s dangerous security environment. Set up MFA for your SSCC account using the same Duo system you use for your NetID today!

Managing Disk Space at the SSCC

Data storage is a major expense for the SSCC, as the performance and reliability required for research data makes “enterprise level” storage much more expensive than the hard drive of your average PC. While many SSCC members use trivial amounts of data storage, if you have a project directory with hundreds of gigabytes of data or more please consider the following:

• Compress large files that are not in active use. Using Compressed Data in Linux has instructions.
• Share large data files among researchers rather than everyone making their own copy.
• Delete intermediate data files that can be reproduced at will, keeping just the raw data and the version of the data you’re currently working on (along with all the code that gets you from one to the other).
• Delete data files that are no longer needed (but only if you’re sure it’s no longer needed).

Very large numbers of small files (10,000+ and some SSCC researchers have millions) use space inefficiently and slow down the file server. Please combine these into compressed archive files if at all possible. Also consider alternatives such as storing your data in a database (learning SQL not required). The SSCC’s statistical consultants can give suggestions.