3. Introduction to the Yen Servers
On-premises GSB resources
As a supplement to everything you can do on your own computer, the GSB has several Linux servers that you can use for your computing needs. These Linux research servers are useful for a variety of tasks, including when you want or need to:
- Run a program over a long period of time and do not want to leave your personal computer on and running
- Run a program that will use a lot of memory (such as when analyzing a large data set)
- Take advantage of parallel processing
- Access software for which you do not have a personal license
- Save files in a place where multiple people can access and work with them
At the GSB, we have a collection of Ubuntu Linux servers (the
yen cluster) specifically for doing your research computing work.
If you are a faculty member, PhD student, post-doc or research fellow, by default you should have access to these servers.
They are administered by the Stanford Research Computing Center (SRCC) and located in Stanford’s data centers.
yenservers are not designed for teaching or course work!
Current cluster configuration
|Server (Host) Name||Physical Processing Cores||CPU Model||Memory|
|yen1.stanford.edu||32||Intel Xeon Gold 6134 @ 3.2 GHz||1.5 TB|
|yen2.stanford.edu||32||Intel Xeon E5-4627 v2 @ 3.3 GHz||1.5 TB|
|yen3.stanford.edu||32||Intel Xeon E5-4627 v2 @ 3.3 GHz||1.5 TB|
|yen4.stanford.edu||32||Intel Xeon E5-4627 v2 @ 3.3 GHz||1.5 TB|
|yen5.stanford.edu||256||AMD EPYC 7H12 @ 2.6 GHz||1 TB|
|yen10.stanford.edu||128||Intel Xeon Gold 6130 @ 2.1 GHz||3 TB|
|yen11.stanford.edu||256||AMD EPYC 7H12 @ 2.6 GHz||1 TB|
Why use the
These servers offer you several advantages over using a laptop or desktop computer.
Let’s use the server
yen3.stanford.edu as an example: this machine has 32 processing cores and 1.5 TB of RAM.
In comparison, my MacBook Pro has 6 cores and 32 GB of RAM. With
yen3, you are able to complete memory- or CPU-intensive work that would overwhelm even the best personal laptop!
Long running jobs
Even when your laptop is capable of doing the job, you may still want to offload that work to the external server. The server can free up resources for your laptop to use for other tasks such as browsing web sites, reading PDF files, working with spreadsheets, and so forth. If your laptop crashes, it’s very convenient for your compute jobs to continue!
Tools like Matlab and Stata are not free for personal use, but are installed and licensed to use on the
The project files and any large output should live on ZFS file system (not in your home). The ZFS capacity is about 612 TB.