Copy
 What's New?
Research Highlights
Events

 

 

New HPC Cluster - Rider

The Research Computing and Cyberinfrastructure (RCCI) group is pleased to announce good progress toward deploying the newest cluster, rider.case.edu. Rider is named in honor of the Case Rough Riders athletic teams, just as RedCat was named for the Western Reserve athletic teams, the Red Cats.

RCCI recently added 36 compute nodes and 8 GPU nodes to Rider, which is based on Red Hat Enterprise Linux version 7 (RHEL7). Each compute node has 24 processors and 96GB of memory. Each GPU node has two NVidia P100 cards, 20 processors and 192GB of memory. At present, RCCI has configured 81 compute and GPU nodes (approximately 1568 cores) in Rider. In addition to the equipment upgrades, RCCI simplified the partitions with simply: batch, gpu and smp partitions.

During the fall, RCCI will gradually migrate compute nodes from RedCat to Rider. On December 1, RCCI plans to begin operating Rider as a production service with "official" migration of users to Rider from RedCat. The hope is to complete the transition no later than June 1, 2018, minimizing the time needed to operate two clusters.

As with previous migrations to new clusters, RCCI needs help in testing the system's functionality and software catalog. If you would like to test the new cluster, please email hpc-support@case.edu and include information about the software and specific node characteristics you would be using. Thanks to those who have already tried out the new cluster and provided positive feedback on its utility.

Back to Top
Want to Contribute?
If you have suggestions for items that you would like to see included in the newsletter, please send them to rcci-newsletter@case.edu or fill out our
Google form.



Reducing Screening Time for DNA Sequences

One of RCCI's experienced cluster users, Aura Perez, M.D., Ph.D., from the School of Medicine, contacted HPC Support to assist in installing an application on RedCat that screens DNA sequences for interspersed repeats and low complexity DNA. HPC Support member Dr. Daniel Balagué Guardia helped Dr. Perez with installing the software.

A few days after the software was installed, Dr. Perez contacted the team again because her jobs were taking longer time to finish than expected. Dr. Balagué investigated the issue and discovered an issue with resource allocation. He scheduled an in-person meeting with Dr. Perez; face-to-face meetings typically offer a better chance to clarify concepts that could be confusing or difficult to explain by email or on a website. 

After this meeting, Dr. Perez and Dr. Balagué changed the resource allocation for her job to match the parallelization required by the code. They were able to slash the computational time by more than half the time it would have taken otherwise.

If you are interested in this topic, please contact RCCI for more information. RCCI continues to look for ways to help optimize research computing involving large-scale computation and complex problem solving. 
 

Back to Top

Events

Title: A Crash Course in Python: Basic Structures
Format: Workshop
Instructor: Daniel Balagué (RCCI & MAMS)
Time:  2:00 P.M. - 4:00 P.M. 
Date: Thursday, October 5, 2017
Location: Toepfer Room
Register Here 

Description: Python is a powerful, multipurpose object-oriented programming language. In this workshop, attendees will learn how Python could be an easy replacement to Matlab or Mathematica in certain situations. Participants will learn the basics of the Python language and its data structures; how to optimize numerical computations using NumPy: "the library for high performance scientific computing"; the concept of "vectorization"; and how to create figures using the Matplotlib library. 

Prerequisites: No Python knowledge required (but the audience is expected to have some basic programming experience). Please have a Python distribution installed (the "anaconda2" Python distribution is recommended https://www.continuum.io/downloads). 

Topics include: 

  • Basic Python: Indentation, Functions Control Flow
  • Python Structures: Lists, Tuples, Sets, Dictionaries
  • Basic Numpy: A Python Library for High Performance Scientific Computing (Multidimensional Arrays, Vectorization and Linear Algebra)
  • Basic Matplotlib: A Powerful Python Library for Visualizing Data

Title: Advancing Visualization with Gnuplot 
Format: Workshop
Instructor: Daniel Balagué (RCCI & MAMS)
Time:  2:00 P.M. - 4:00 P.M. 
Date: Thursday, October 19, 2017
Location: Toepfer Room
Register Here 
 
Description: Gnuplot is an open source readily available plotting tool on Windows, Linux and Mac platforms. It can be integrated into scripts easily to produce clean-looking initial graphs or high quality plots for research publications. The workshop will cover basic syntax, the basic structure of Gnuplot's data files, plotting functions and data files, and scripting.
 
Prerequisites: No previous knowledge of Gnuplot is required. Gnuplot is a scripting language, so being familiar with the terminal would be useful.
 
Topics include:

  • Ask Gnuplot for Help
  • Plot Simple Functions
  • Understand Some of the Basic Options for 2D and 3D Plotting
  • Understand Data File Structures
  • Plot Different Types of Data Files
  • Know Where to Find References and More Examples
     
Copyright © 2017, All rights reserved.
Case Western Reserve University

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list