Version 19 (modified by 9 years ago) ( diff ) | ,
---|
About Cypress
The Tulane University High Performance Computing system, code named “Cypress,” is one of the fastest computers in the world. In November 2014, the TOP500 ranked Cypress as the 271st supercomputer site in the world. It is the 3rd fastest among private American universities and 10th fastest among across all American universities.
Cypress is located in the Tulane datacenter in Downtown New Orleans. It is composed of Dell state-of-the-art technology, using Intel Ivy Bridge CPU, Intel Xeon Phi Coprocessor, and Lustre file system running on Dell storage with Intel Enterprise Edition for Lustre (IEEL) technology. The components of Cypress were carefully selected to provide consistent performance over a wide range of scientific applications. Cypress delivers a peak general capability of 369Teraflops and incorporates 11.84TB of fast main memory with 1.476 Petabyte of parallel-distributed file system Lustre high-performance storage, 480TB of which is dedicated to Hadoop with IEEL (Intel Enterprise Edition for Lustre) high performance storage for use by the Hadoop data analytics queue.
Node Type | C8220X/E5-2680 v2 & Xeon Phi 7120P |
# of Nodes | 124 |
CPUs per node | 2 |
Total E5 2680 v2 Cores | 2,480 |
Total Xeon Phi cores | 15,128 |
Tflops CPUs only (theoretical) | 69.44 |
Tflops system peak w/ Phi (theoretical) | 369 |
Memory per node | 12 nodes: 256GB each - 25 nodes: 128GB each - 87 nodes: 64GB each |
Total Memory | 11.84TB |
Storage | 1.476PB expandable to 4PB |
Getting an account
All accounts must be associated with a Tulane faculty member who is the Principal Investigator (PI) of his/her research group. Please have your PI contact us to establish a group on Cypress. (If some members of your group have Cypress accounts, your group is already established.) After your group is established, visit Tulane Service Now at:
https://tulane.service-now.com/
to request an account. After logging in with your Tulane credentials, select "Order Things", then "Accounts", and finally "Cypress HPC Request" and fill out the form.
Getting Started on Cypress
Follow the links below for information on obtaining an account, and instructions on using Cypress.
Working on a remote Unix/Linux system
- SSH, terminal software
- Unix/Linux commands
- File Editing Software
- File transfer to/from your desktop/laptop
- Working on a remote Unix/Linux Cluster
- Module command
- How to submit a job
- Running interactive jobs (idev)
Scripting on Cypress
How to run a Matlab code on Cypress
How to run R code on Cypress
How to run Python Cypress
Coding on Cypress
Coding
Other Topics
Version control
Compilation, debugging, and optimization
- Makefiles
- Code Debugging
- Code Profiling
- gdb
- valgrind
- Intel Parallel Studio XE (Inspector, Advisor, and Vtune)