Tag Archives: high performance computing

Lecture: High Performance Computing, Technologies and Opportunities — Wednesday, May 8

Posted on by

This talk will provide a high-level introduction to high-performance computing. The first half of the talk will be an overview of high-performance computing, including common parallel programming models and how best to run them on high-performance computers. Both traditional CPU and GPU (CUDA, OpenGL) architectural models will be covered. In the second half, we will review some resources for high-performance computing that are available now, at UM and beyond, including UM’s Flux high-performance computing cluster and the XSEDE virtual cyber infrastructure. Attendees will come away from this talk with a broad overview of, and concrete next steps to take in, engaging with high-performance computing.

Time: 3-5 p.m., Wednesday, May 8

Location: Screening Room, 2160 Shapiro Library, 919 South University, Central Campus

Session level: Beginner

Sponsor(s): Advanced Research Computing at U-M (ARC), LSA Information Technology Advocacy and Research Support

Presenter: Charles Antonelli, LSA Information Technology

For more information and to register, please visit https://ttc.iss.lsa.umich.edu/ttc/sessions/high-performance-computing-technologies-and-opportunities/

U Nebraska-Lincoln Holland Computing Center Applications Specialist Opening

Posted on by

This position is responsible for contributing to a broad mission of the Holland Computing Center (HCC), promoting the use of high performance cyberinfrastructure and a wide variety of research projects requiring high performance and/or high throughput computing, storage, and other cyberinfrastructure. Responsible for working with researchers and supercomputer system administration staff and students to port, produce and optimize code, and facilitate and organize grant proposals related to, and supportive of, the facilities of the HCC. Criminal background check will be conducted. Excellent benefits including staff/dependent scholarship program.

Applicant review begins March 6th, so please submit your application prior to March 6th. View requisition S_130092 at https://employment.unl.edu/postings/search for details and to apply. UNL is committed to a pluralistic campus community through affirmative action, equal opportunity, work-life balance, and dual careers. Masters degree in Computer Science or field related to Scientific Computing, plus two years experience in system administration and scientific computing required. PhD in Computer Science or field related to Scientific Computing; experience in Grant Writing or Computational Research preferred.

For further details see https://employment.unl.edu/postings/34556

 

Florida State University Research Computing Center Application Support Specialist Job Opening

Posted on by

 

The Florida State University recently restructured its shared-HPC facility to better reflect its broad mission and to improve support for computational and data intensive research.  The name of the new organization is the FSU Research Computing Center.  We now seek talented individuals to join a team of five other professionals to support a broad portfolio of research computing related services.  A brief description of the position is given below.  Please contact Jim Wilgenbusch (wilgenbusch@fsu.edu) if you have questions about this position. This is a two year time limited term position, and applications need to be received by June 10, 2013.

 

Application Support Specialist:  Provides application support to a diverse set of research areas. This support will include writing scripts and programs, compiling, testing, documenting, and installing software packages, answering general application questions, and developing databases where needed.  The staff member will need to collaborate with faculty on issues related to research and data analysis and mentor and train student workers assigned to application support tasks.  Candidates with a Ph.D. degree in an appropriate area of specialization and demonstrated experience with a compiled language, such as C, Fortran, or C++, and MPI will be preferred.  You can view and apply for this job at: https://jobs.omni.fsu.edu/psp/sprdhr_er/EMPLOYEE/HRMS/c/HRS_HRS.HRS_APP_SCHJOB.GBL?JobSeekers_ApplyForAJob=35365&Seq=1

 

Supercomputing in Plain English: A High Performance Computing Workshop Series, Starting January 22

Posted on by

Supercomputing in Plain English (SiPE), Spring 2013
Available via free videoconferencing

Tuesdays starting Jan 22 2013, 2:00 pm Central Time
(4:00pm Atlantic, 3:00pm Eastern, 1:00pm Mountain, 12:00noon Pacific)

These workshops focus on fundamental issues of HPC as they relate to Computational and Data-enabled Science & Engineering, including:

* the storage hierarchy;
* instruction-level parallelism;
* high performance compilers;
* shared memory parallelism (e.g., OpenMP);
* distributed parallelism (e.g., MPI);
* HPC application types and parallel paradigms;
* multicore optimization;
* high throughput computing;
* GPGPU computing;
* scientific and I/O libraries;
* scientific visualization.

The key philosophy of the SiPE workshops is that an HPC-based code should be maintainable, extensible and, most especially, portable across platforms, and should be sufficiently flexible that it can adapt to, and adopt, emerging HPC paradigms.
SiPE is targeted at an audience of not just computer scientists but especially scientists and engineers, including a mixture of undergraduates, graduate students, faculty and staff.

Prerequisites:
* For the presentations:
– at least one semester of programming experience in any of C, C++ or Fortran, recently.
* For the HPC exercises:
– at least one semester of programming experience in any of C, C++ or Fortran, recently;
– at least one semester of experience with any Unix-like operating system (could be Linux but doesn’t have to be), recently (if someone doesn’t have this, it’s not a crisis, because the basics are covered in the first exercise).

Registration is now open! Send e-mail to hneeman@ou.edu.
Videoconferencing: details to be announced
Streaming video recordings: details to be announced
Details at: http://www.oscer.ou.edu/education/

NOTE: Please plan to participate every Tuesday Jan 22 – Apr 30 2013 except Tue March 19.

So far, the SiPE workshops have reached roughly 1000 people at 166 institutions, agencies, companies and organizations in 42 US states and territories and 5 other countries:

* 125 academic institutions;
* 18 government agencies;
* 16 private firms;
* 7 not-for-profit organizations.

HPC 101 Session – Learn How to use the Flux Computing Cluster, January 23

Posted on by

Here’s an opportunity for Flux cluster users and potential users to learn more about the Flux computing cluster, creating and submitting jobs, understanding the available tools, and avoiding common pitfalls.  This will be a hands-on experience, in which you will log in to the cluster and work with jobs.  Basic Linux command-line skills are a pre-requisite, as the Flux cluster is implemented on Linux.

The class will be offered Wednesday, January 23, from 1 – 5 PM in Room B745 East Hall.  Full details at http://unix.lsa.umich.edu/HPC101/ .
Course registration link at https://www.engin.umich.edu/form/hpc101_20130123 .

We plan on teaching this course again next term, so folks can take it later if they can’t make it this time.

We think graduate students will want to attend, and faculty and admins are certainly welcome as well.

Learn to Use Flux – Sept 19

Posted on by

Here’s an opportunity for Flux cluster users and potential users to learn more about the Flux computing cluster, creating and submitting jobs, understanding the available tools, and avoiding common pitfalls.  This will be a hands-on experience, in which you will log in to the cluster and work with jobs.  Basic Linux command-line skills are a pre-requisite, as the Flux cluster is implemented on Linux.  Full details are at unix.lsa.umich.edu/HPC101/.

Date:  Wednesday, September 19
Time:  1 – 5 PM
Location:  Room B745 East Hall

This class is appropriate for researchers who are using or contemplating using Flux for their research and graduate students. Faculty and admins are welcome as well. This course is taught every term, so those who are interestesd can take it later if they can’t make it this time.

Call for Proposals – Great Lakes Consortium for Petascale Computation

Posted on by

The Great Lakes Consortium for Petascale Computation (GLCPC) has issued a Call for Proposals (CFP) for allocations on the Blue Waters High Performance Computing System. The GLCPC has been allocated 50 million core hours annually as part of the Blue Waters project. This allocation provides the GLCPC member institutions with an unprecedented opportunity to advance their programs in computation-, data-, and visualization-intensive research and education.

This CFP describes the process for submitting a proposal to the GLCPC Allocations Committee for allocations on the Blue Waters system. Details on the Blue Waters system can be found at www.ncsa.illinois.edu/BlueWaters/. Given the unprecedented scale and innovative architecture of the Blue Waters System, we are seeking proposals that focus on its scale and unique capabilities. Thus, projects that could be completed on one of the other NSF sponsored systems as part of the current XSEDE program are not encouraged for GLCPC Blue Waters Allocations.

The GLCPC is seeking innovative proposals that fall into four categories:

1) Scaling studies: The scaling of codes which will operate efficiently on large numbers of parallel processors presents a number of challenges. Therefore, projects of particular interest include those that optimize and/or scale community codes to very large scales. Examples include scaling of multilevel parallel applications (MPI+OpenMP), accelerators (CUDA, OpenACC or OpenCL), I/O and Data intensive applications, or novel communication topologies.

2) Multi-GLCPC-institutional projects addressing focused scientific projects. An example might be a Great Lakes Ecosystems Modeling initiative (Digital Great Lakes).

3) Proposals for applications well-suited for the BW system architecture.

4) Proposals from non-traditional and underserved communities.

The GLCPC Allocations Committee anticipates 3-8 projects/allocations annually; consequently, the smallest project is expected to be approximately five million core hours, which is roughly the same as dedicated use of a 4-core, 1280-node cluster. We note that applications at this scale will require development efforts as well as different phases, such as: tuning and development; some smaller runs; large “production runs”; and then post processing; but, all will be at scales beyond other available large resources. The Blue Waters system is expected to be operational and available for GLCPC allocation use by March 2013.

Deadline: GLCPC allocation proposals will be accepted through October 31, 2012. The proposal review process is expected to be complete by early January 2013. Allocations awarded through GLCPC will expire one year from time of award. To make your allocation request, please submit a written proposal, not to exceed five pages (in PDF format).

Eligibility: Only Principal Investigators affiliated with an institutional member of the Great Lakes Consortium for Petascale Computation are eligible to submit a GLCPC allocations proposal. Access to the Blue Waters system will be governed by the Blue Waters Terms of Use. Questions about this CFP should be addressed to: allocations-glcpc@ncsa.illinois.edu.

Proposal Content (not to exceed 5 pages):

Target Problem: A description of the specific research question(s) that the resources requested will
be used to answer and the scientific and societal impact of the proposed work. Include an explanation of why a petascale resource of the leading-edge capability that Blue Waters represents is necessary to address this research.

Team Members: Name, title, affiliation and contact information for each member of requesting
project team.

Description of codes: Describe the structure of the application codes that you intend to use. These
may either currently exist, but might require enhancement, or they may be in development. Please include details about the algorithms involved and the approach that you intend to use to ensure that the code scales effectively on the Blue Waters architecture.

Please include descriptions of how your code(s) will use each of the major system elements:

  • the memory hierarchy,
  • the communications
  • network,
  • the computational elements, and the
  • I/O subsystem.

Describe how you intend to analyze the output resulting from your use of Blue Waters. IMPORTANT: Please describe any run-time libraries or special system software or program development environment features that you will require and the types of graphics support that you would find most useful.

Porting, testing and scalability plans: Describe the current state of readiness of the application codes that you intend to use and your plans for developing these to the point where they are ready to run in production mode on the Blue Waters system. Estimate the type of consulting help that you will need from the Blue Waters project team to port and scale your application(s) to the Blue Waters system.

Resources Required: Describe the Blue Waters resources required to complete research on the
Target Problem. This description should include the number of system nodes needed for your runs, the anticipated actual memory usage, the expected numbers of each major class of arithmetic and logical operation, the expected numbers of local and remote memory accesses, the total number of core-hours required, the anticipated input and output requirements, the amount of data that you anticipate transferring to or from the Blue Waters enclave, the amount and type of storage required
and any other system resource needs that you anticipate.

Support Plan: Allocations will be made with the understanding that the support model is explicitly
targeted for experienced, advanced users. NCSA does not have the staff to provide a broad range of support or introductory services. Proposals must describe the institutional structure that will be used to support the proposed work. Institutional support staff will be responsible for triage of user problems and basic user support. Blue Waters experts will help train the support staff at the proposing institutions, and will also help triage problems.

Source(s) of Funding: Please identify the source(s), amount(s) and duration of existing funding that
is supporting the development of your petascale application. If not currently funded, please describe how you intend to support any required development work.

Additional Required Documents (not included in 5 page proposal limit):

  • NSF formatted 2 page bios of project team PI and co-PIs
  • References cited

Questions about the CFP can be addressed to the allocations committee at allocations-glcpc@ncsa.illinois.edu.

HUBbub 2012 – Sept 24-25

Posted on by

HUBbub 2012: The HUBzero Conference, will take place September 24-25, 2012, in Indianapolis, Indiana. Come to the conference to eee how the unique HUBzero solution has empowered a wide spectrum of projects in nanotechnology, healthcare research, and other areas of engineering and science. You will learn through hands-on tutorials:

  • how to set up your own hub using HUBzero’s open source software,
  • how to create and publish scientific tools on your hub,
  • how to connect the tools to computing clusters and other Grid resources, and
  • how to add new capabilities to the platform.

For more information and to register, go to hubzero.org/hubbub.

HUBzero is a platform for creating web sites that support scientific research, education, and collaboration. HUBzero was released as open source during HUBbub 2010 and a new version will be released this year at HUBbub 2011. The HUBzero Platform is the basis of nanoHUB.org and 25 other sites, delivering hundreds of scientific tools and seminars to more than 450,000 visitors each year. HUBzero is supported by a consortium of universities including Purdue, Indiana, Clemson, and Wisconsin.

First International Conference on Space, Time, and CyberGIS – Aug 6-9

Posted on by

The first International Conference on Space, Time, and CyberGIS (CyberGIS’12) will take place August 6-9, 2012, in Urbana-Champaign, Illinois. This conference aims to bring together researchers, educators, and practitioners for sharing and synergistically advancing the state of the art of CyberGIS and Space-Time Analysis/Modeling/Synthesis; and to foster international collaboration and cooperation in this area.

The CyberInfrastructure and Geospatial Information Laboratory (CIGI) and National Center for Supercomputing Applications (NCSA) will host the conference on the campus of the University of Illinois at Urbana-Champaign. CIGI researches and develops cutting-edge cyberinfrastructure to advance geospatial sciences and technologies while NCSA has been a world leader in advancing and supporting cyberinfrastructure, and in particular high-performance computing and scientific visualization, for more than 25 years.

For more information and to register, please visit cigi.illinois.edu/cybergis12/.

New NASA Supercomputer Facility Set to Advance Earth Research

Posted on by

WASHINGTON — NASA soon will open a new chapter of discovery using enhanced Landsat Earth-observing data in a state-of-the-art, high-performance computing and data access facility called NASA Earth Exchange (NEX). This new facility is a virtual laboratory that will allow scientists to tackle global Earth science challenges with global high-resolution satellite observations.

After extensive development and testing, NASA is making the NEX facility available to the research community for further research and development. With NASA’s state-of-the-art supercomputing capacity, researchers can use NEX to explore and analyze large Earth science data sets in hours rather than months. Scientists can produce complex, interdisciplinary studies of world phenomena and share their findings instantly on the NEX platform.

“Because of the large volume of high-resolution Landsat data, scientists who wanted to study the planet as a whole prior to NEX needed to invest tremendous amounts of time and effort to develop high-end computational methods rather than focus on important scientific problems,” said Tsengdar Lee, high-end computing program manager at NASA Headquarters in Washington. “NEX greatly simplifies researchers’ access to and analysis of high-resolution data like Landsat.”

This new facility boasts a large collection of global data sets and analysis tools from NASA and other agencies, including surface weather records, topography, soils, land cover and global climate simulations. Using NEX, scientists now can fit Landsat scenes together like a giant jigsaw puzzle to create snapshots of global vegetation patterns containing more than a half-trillion pixels in less than 10 hours. These global vegetation products, referred to as the Normalized Difference Vegetation Index, complement the more standard products from the Moderate Resolution Imaging Spectroradiometer on NASA’s Aqua satellite but with 10 times higher resolution.

“The science community is under increasing pressure not only to study recent and projected changes in climate that likely impact our global environment and natural resources, but also to design solutions to mitigate, or cope, with the likely impacts,” said Rama Nemani, a senior Earth scientist at NASA’s Ames Research Center in Moffett Field, Calif. “We want to change the research paradigm by bringing large data holdings and supercomputing capabilities together, so researchers have everything they need in one place.”

Developed by a team at Ames, NEX combines Earth-system modeling, remote-sensing data from NASA and other agencies, and a scientific social networking platform to deliver a complete research environment. Users can explore and analyze large Earth science data sets, run and share modeling algorithms, collaborate on new or existing projects and exchange workflows and results within and among other science communities.

Scientists believe costs and time associated with research development may be reduced significantly by allowing NEX members to collaborate instantly in this type of large-scale supercomputing work environment. For example, NEX may relieve researchers from redundantly retrieving and integrating data sets and building modeling analysis codes.

NEX uses Landsat data, which constitute a large collection of images collected over 40 years by a series of satellite sensors. The enhanced collection of Landsat data gives scientists the opportunity to study and understand changes on a planetary scale, looking at one-quarter acre at a time.

NASA, in cooperation with the Interior Department and its science agency, the U.S. Geological Survey, launched the first Landsat satellite in 1972. The resulting 40-year archive of Earth observations from the Landsat fleet supports the improvement of human and environmental health, biodiversity, energy and water management, urban planning, disaster recovery and crop monitoring. The Landsat program is jointly managed by NASA and the Interior Department.

For more information about the NASA Earth Exchange, visit: nex.arc.nasa.gov.

For information about the NASA Advanced Supercomputing facility, visit: www.nas.nasa.gov.

For more information about NASA, visit: www.nasa.gov.

See see this press release, visit: www.nasa.gov/home/hqnews/2012/jul/HQ_12-248_NEX_Landsat_Available.html.