Yes, in January 2019, we started offering accounts on R Studio Servers. An R Studio Server is a computer on the Internet, running a webpage offering the R Studio Environment. Its just like running R Studio on your own laptop or desktop. The advantage to using an R Studio Server include:
- More Memory
- More Processor Cores
- You can generally run jobs longer and they won't stop if you close your laptop
- Configured to support the libraries you use
As we gain experience offering these new services and servers we will be iterating on improving the setup, configuration, access and documentation. Please provide us any feedback so we can improve the service.
Basic usage tips:
- you'll generally access the server over port 8787 (we'll send you access information) when you request an account.
- The account and password that you'll use won't match your Bren Desktop or UCSBnetID, its another account (sorry).
- You can access the service from on campus addresses (or you can vpn from home).
- When you login, you'll login over http, which isn't secure, but RStudio encrypts the password before it sends it, so its safe.
- You can transfer files via secure copy (using tools such as fetch, winscp, or command line tools such as scp).
- We aren't backing up this machine, so make sure you are using github for your code.
The two primary servers we have now include one in the Amazon AWS Cloud, that is meant for instructional access (when a student can't configure their laptop properly with R) and a computational server intended for Faculty, PhD and Bren Research efforts. The computational server has 256 GB of memory, 20 cores, and 2 TB of storage. If you have large datasets, contact us, and we can add storage. We also have Anaconda installed on this server for python datascience computation.
Things we will be investigating:
- How to access / copy files from cloud storage (like google drive using rclone)
- Helping you use packrat to package any libraries / versions that are different than what we have on the server.