Copy
View this email in your browser
Research Computing + Cyberinfrastructure

OnDemand HPC Web Portal

Ondemand.case.edu continues to gain popularity for its ease of use and access to Rider and Markov cluster resources via the web browser.  The web portal provides productivity “dashboard” apps that feature file browsing, a terminal, and job submission management and monitoring. Additionally, interactive applications provide a web-interface to:
  • Jupyter
  • Rstudio,
  • A full desktop environment, with choice of Mate, Xfce4, or KDE4 window managers
A beta-deployment Jupyter with Tensorflow interactive application is available on GPU nodes and makes use of software maintained on the HPC cluster.

In order to access the portal, you need to have an existing HPC account, and can sign on to applications using Case SSO (Single Sign-On) with Duo two-factor authentication. VPN access is not required to access OnDemand unlike all other HPC services.

All interactive applications provide the choice of allocating GPU resources.

See our quick start guides through the HPC documentation page

Additional HPC Compute Resources

RCCI has added significant resources to both the Rider and Markov clusters since Spring 2020. To support users of the Rider cluster, an additional 24 standard compute nodes with 40 processors and 192 GB of memory have been added. GPU capacity has been increased by adding an additional 10 GPU nodes with 24 processors and 192 GB of memory each. GPU nodes are configured with either 2 or 4 Nvidia V100 GPU cards. 

The Markov cluster, that is configured for mostly class use, was expanded by the addition of 28 standard compute nodes, each with 40 processors and 192GB of memory. This complements the 20 GPU nodes that have been available since last year.

Faculty who would like to use cyberinfrastructure including the Markov cluster for teaching can apply for resources by filling out this short form. Additional information can be found on our UTech website.

Compilers For Software Development

Intel 2020 Parallel Studio XE Cluster Developers License

Intel Parallel Studio XE is a software development product from Intel that facilitates native code development on Windows, macOS and Linux in C++ and Fortran for parallel computing. Parallel programming enables software programs to take advantage of multi-core processors from Intel and other processor vendors. 

The module 'intel-20.1' sets environment variables necessary to access the parallel studio components on both Rider and Markov HPC clusters. 
 

PGI 2020 Professional License

PGI Professional Edition is a perpetual license that provides PGI Fortran, C - C++ compilers, and associated tools for multicore CPUs and NVIDIA Tesla GPUs, including all OpenACC, OpenMP and CUDA Fortran features. The license and compiler enables development of performance-portable HPC applications with uniform source code across the most widely used parallel processors and systems. PGI couples effectively with the Nvidia HPC Software Developers Kit for GPU and multi-core processing.
 

The module 'pgi/20.1' updates your environment to use PGI components on both Rider and Markov HPC clusters.

Support And Consultation Services

RCCI continues to support the campus community through the use of Zoom and other communications tools, in lieu of in-person consultation. Please reach out if you need help with access or would like to have a conversation on ways to optimize your current research activities. We are committed to continuing the education and research mission of the university. You can view our services and resources directly on the UTech website.

Pre-Award Resources

Funding agencies often require detailed language about the computational resources that will be used as part of a research project. To facilitate the application process, RCCI has created proposal templates detailing the design and capabilities of resources such as the High Performance Computing Cluster.  The proposal templates are available in Box for review.

If you are planning to apply for external funding and would like assistance with areas such as data management or resource budgeting contact RCCI at rcci-support@case.edu

Upcoming RCCI Workshops

 
High Performance Computing Introduction
September 10th, 9:00 AM - Noon   
Zoom webinar - Register Now 

RCCI announces our start of semester training for those needing to improve their familiarity with the CWRU HPC.

This virtual event is geared towards new users to the CWRU HPC cluster, presenting introductory materials in lecture with demonstrations. Our goal is to provide a grounding in the concepts necessary to conduct work in the HPC environment, and to make clear the methods for connecting to the HPC from personal and lab computers. We will provide an overview of the cluster hardware, software and policies, as well as an introduction to the Linux command line, scheduling computing resources, data management and data transfer methods. 

To see the program, visit our Events page
 

Working with OnDemand 
September 17th, 10a.m. - 11a.m.    
Zoom webinar - Register Now

Working with OnDemand HPC Portal will introduce the Open-OnDemand (OOD) web portal for accessing CWRU HPC resources. This includes signing in via Case SSO (Single Sign-On), obtaining a terminal, monitoring and submitting jobs, and, managing files on the cluster. In addition, attendees will gain experience with allocating resources for web-based Jupyter notebook and Desktop interactive sessions

To learn more about our workshops and other training opportunities check out our RCCI Workshops site.

Copyright © 2020, 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