ÐÇ¿Õ´«Ã½

Availability + Service Expectations

• The cluster is currently available to Case faculty for academic and research purposes.

• RCCI-managed services include provisioning account access to our computational resources, assistance in optimizing and providing answers to questions about use of the resource.

• HPC Guest Services: Guest level membership access is available for faculty members that want to try out HPC services for a period of one year.

  • 36 hours of walltime per job
  • 24 processors in simultaneous use by jobs by group
  • 12 jobs permitted to run simultaneously within a guest group
  • A soft disk quota of 150 GB and a hard disk quota of 260 GB
  • Basic consultation services

• HPC Member Services:

  • 320 hours of walltime per job
  • Total number of processors in simultaneous use by jobs within a member group is expanded to four times the size of the membership share of that group
  • No limit on the total number of simultaneous jobs for any member account or account group
  • A soft disk quota of 700 GB and a hard disk quota of 920 GB.
  • Expanded consultation services
  • Application installation and administration

Current Hardware + Operating Systems

• Dell PowerEdge servers with Intel Xeon processors

• GPU nodes for higher end graphics processing

• SMP nodes for intensive memory needs

• Compute nodes for general purpose tasks

• Cluster nodes run Red Hat Enterprise Linux

Potential Uses of Higher Performance Computing Cluster:

• Computational chemistry to find new drugs or material, or to study molecular interaction and its effect

• Computational fluid dynamics to study material stability, fire spread, explosion

• Genomics study to look at the effects varying genes on human health and behavior

• For an overview of ÐÇ¿Õ´«Ã½ faculty research utilizing HPC resources, please visit our page.

Associated RCCI Resources + Services

Services

• Consultation Services - RCCI can provide the necessary information about the existing Core facility (boilerplate language), suggest the appropriate computational hardware and storage for the proposal, including the necessary quotes, and execute the proposed computational plan upon the award being funded.

• Big Data Cluster - RCCI can provide a separate cluster to work on Big Data by constructing the cluster with Cloudera Hadoop. Applications in this environment include: HDFS/Yarn, HBase, Spark, Hive, etc.

• Secured Research Environment - RCCI provides a secured research environment to keep and to analyze protected data under PII, FISMA and HIPAA. Further information can be obtained by clicking on the link above.

Resources

• External Resources - While our resource serves well for many research uses, it is not of adequate size to handle larger computational tasks that may require several hundred or several thousand simultaneous processors. For those larger jobs, there are other resources available to ÐÇ¿Õ´«Ã½ researchers that are both free and commercially available.

• Visualization Resources - Case UTech research computing supports visualization activities for researchers throughout the Case community. In addition, research computing has computing systems that provide the means to produce, explore or share visualization materials.

Who manages the HPC Service?

• The HPC Service Manager is Hadrian Djohari. He can be reached at hpc-support@case.edu.

How can I get the HPC Service?

• After reviewing the Membership Options above, eligible faculty should complete the . Contact hpc-support@case.edu with any questions regarding the account application process.

How do I cite the use HPC resources in my research work?

• We ask that you acknowledge the use of the HPC system in your publications and presentations as
  • "This work made use of the High Performance Computing Resource in the Core Facility for Advanced Research Computing at ÐÇ¿Õ´«Ã½."
• Please email references of your published research papers or poster presentations at <IMAGE OF hpc-support@case.edu>

Can I view the current status of the HPC cluster?

• Those with an HPC cluster account and password can view the current status of the cluster . Note: VPN is required to access this service from off-campus.

How is the HPC sustained and updated?

• HPC services are funded through a mixture of UTech operating funds and faculty membership subscriptions. Subscriptions can be paid for through direct investment in HPC equipment or an annual charge per the amount of resources a researcher utilizes. UTech utilizes these funds to pay for licensing and infrastructure upgrades as necessary.

Where can I find more information about UTech Research Computing and the HPC?

In addition to the Additional HPC Information section below, please view our Research Computing Brochure.

• These sites provide you with pointers that gives good insight on HPC servers and storage on the fly. Knowledge on type of nodes and their attributes helps you to ask for appropriate resources for your jobs whereas familiarity on individual and group quota keeps your jobs seamlessly running in the cluster without affecting your group members.

• These pages answer questions related to accessing HPC System.

• - This page answers questions related to visual/GUI access to the HPC.

• - In this page you will find some tutorials and "how to" guides that will help you with some common tasks in HPC.

• - This page covers things to take into account before you run your HPC job.

• - This page provides a list of software applications available on HPC System.

• - This pages contains links to basic information about Linux, Parallel computing, Parallel Programs, debugging, compiling, and linking, Software installation, special job submission, and more.