High Performance Computing

  • High Performance Computing

    HPC Announcements and News

    ATS/IDRE-provided High Performance Computing Services

    Contact: atshpc@ucla.edu

       

      Contents:

       

      Hoffman2 Shared Cluster

      The Hoffman2 Shared Cluster, as the heart of the ATS/IDRE Shared Cluster Hosting Program, is logically organized into several component clusters that have been optimized for different research needs. The components include: the Campus General Purpose Cluster and the Research Shared Virtual Cluster made up from Base cores provided by ATS/IDRE and Contributed cores provided by researchers.

      By prior special arangement with ATS/IDRE, runs utilizing a large part of Hoffman2 can be arranged. Please contact atshpc@ucla.edu for further information.

       

      XSEDE Computing Resources

      UCLA-ATS/IDRE is part of XSEDE Campus Champion Program: The XSEDE Campus Champion, your local source of knowledge about national high performance computing opportunities and resources, can assist in the following ways:

      • Provide you with information about high performance computing resources available to researchers and educators, including resources available from the XSEDE
      • Assist you to quickly get access to allocations of high performance computing resources
      • Facilitate awareness and training workshops for you about the use of high performance computing resou5rces and services
      • Provide you with contacts within the high performance computing community for quick problem resolution
      • Provide you with timely information on new resources, services, and offerings

      To learn more about available high performance computing opportunities, contact UCLA's XSEDE Campus Champion.

       

      Additional Computing Resources

      NSF and DOE centers: NERSC, NASA, ALCC and INCITE.

      SDSC: Triton Affiliates and Partners Program (TAPP)

      ATS/IDRE-HPC team can assist with the scaling issues and studies that may help predict the timing on large computing resources. Please contact atshpc@ucla.edu with a brief proposal. 

       

      HPC Consulting (by prior appointment through mail to atshpc@ucla.edu)

      • Multicore/multiprocessor programming
      • High-end visualization
      • Individual code clinics
      • Scaling issues and analysis of parallel codes
      • Optimization of serial codes
      • Use of efficient serial and/or parallel algorithms
      • Parallelization or porting on different platforms
      • Debugging
      • Profiling

       

      IDRE-HPC Project Opportunities

      Focus Projects: Faculty and researchers working on questions involving high-performance computing are eligible to receive IDRE staff and resource support through the annual IDRE Call for Focused Research Projects.

       

      Education/Classes

      • High Performance Computing Classes include regularly scheduled classes in Hoffman2 cluster usage, shared-memory thread programming with OpenMP, distributed-memory message-passing programming with MPI, and GPU computing with CUDA.
      • Online class: Introduction to Parallel Computing.
      • We can present a class at your request in your department to groups of 5 or more people. We will assist you and/or guest lecture in any regular University class.


      Shared Cluster Hosting Program- Hoffman2 Shared Cluster

      ATS/IDRE runs the Hoffman2 cluster in accordance with the ATS/IDRE's Shared Cluster Hosting Program. The Hoffman2 Shared Cluster provides campus researchers a way to effectively manage limited high-end data center space on campus. It both maximizes the number of supported customers, and minimizes the labor to support a given cluster, while at the same time providing a rich and robust set of hardware, software, application and support services. Part of the Hoffman2 Cluster is set aside as a free resource for computation by general campus users.

      If you are interested in participation, please contact atshpc@ucla.edu for further information.


      UCLA Grid and UC Grid Portals

      UCLA Grid and UC Grid portals provide web-browser access to computational clusters on the UCLA campus and system-wide at the University of California respectively.


      ATS/IDRE-Developed and Distributed Software

      UCLA Grid Portal: A Grid is a collection of independently owned and administered resources which have been joined together by a software and hardware infrastructure that interacts with the resources and the users of the resources to provide coordinated dynamic resource sharing. The UCLA Grid Portal (UGP) provides a single web interface to all the computational clusters in a Grid. It is used to run the UCLA Grid Portal and Grid Portals at other University of California campuses which are participating in the UC Grid.

      Download the UGP software and access the UC Grid Wiki

      Queue Scripts: To submit a batch job to a scheduler such as the Sun Grid Engine, the scheduler used to manage batch jobs on ATS/IDRE-hosted clusters, users normally have to know and use the scheduler commands and, for each job to be submitted, prepare a command file of scheduler commands for that job.

      The ATS/IDRE-developed queue scripts autmatically create scheduler command files for user jobs and handle the interface between the user and the scheduler. Not only do the queue scripts simplify the submitting of jobs for the users, but use of the queue scripts increases job success by reducing mistakes.