Skip to main content

OLCF CENTER FOR ACCELERATED APPLICATION READINESS

Proposals for CAAR partnership projects must be submitted by February 20, 2015, to [email protected].

The Oak Ridge Leadership Computing Facility (OLCF) is now accepting proposals for Center for Accelerated Application Readiness (CAAR) partnership projects. The CAAR will establish eight partnership teams to prepare computational science or engineering applications for highly effective use on the OLCF system to be named Summit that will become available to users at the OLCF in 2018. The architecture of Summit will consist of nodes with multiple IBM Power-9 CPUs and NVIDIA Volta GPU accelerators, using a coherent memory space that includes high bandwidth memory (HBM) on the GPUs and a high speed NVLink interconnect between the POWER9 CPU and Volta GPUs. Internode communication will be through a Mellanox InfiniBand EDR interconnect. The peak performance of this system is expected to be five to ten times that of Titan.

Each of the CAAR projects will consist of a three-year Application Readiness phase (2015-17) in which the code refactoring and porting work will take place and an Early Science phase (2018) for tuning of the code to the Summit architecture and demonstration of the application through a scientific grand-challenge project.

In addition to application readiness for Summit, performance portability to other architectures is an important consideration, and the CAAR is collaborating with the Argonne Leadership Computing Facility (ALCF) and the National Energy Research Supercomputing Center (NERSC) to enhance application portability across their respective architectures.

The partnership teams, consisting of the core developers of the application and staff from the OLCF that will be assigned to the project, will receive support from the IBM/NVIDIA Center of Excellence at Oak Ridge National Laboratory, and have access to computational resources including Titan at OLCF, Mira at ALCF and Edison and Cori at NERSC, early delivery systems and Summit as they become available.

 

Partnership team responsibilities

  • Develop a technical application porting and performance improvement plan with reviewable milestones for the Application Readiness phase of the project
  • Develop a management plan with clear description of responsibilities of the CAAR team consisting of the core application developers, the Scientific Computing staff member assigned by the OLCF and the OLCF postdoctoral fellow, that will carry out the code optimization, refactoring, testing and profiling of the application
  • Develop a compelling scientific grand-challenge campaign for the Early Science phase of the project
  • Assign an application scientist who, together with the CAAR team, will carry out the Early Science campaign
  • Prepare the necessary documentation for semi-annual reviews of achieved milestones, and intermediate and final reports

 

Partnership resources

  • The core development team of the application, with a stated level of effort dedicated to the partnership
  • An ORNL Scientific Computing staff member, who will partner with the core application development team to jointly carry out the code profiling and optimization tasks, The OLCF commits a minimum of a third FTE per year to the partnership
  • A full-time postdoctoral fellow, located and mentored at the OLCF, who will engage with the CAAR team for code profiling, optimization and execution of the science challenge
  • Allocation of compute resources on Titan
  • Allocation of compute resources at the ALCF and at NERSC to enable performance portability to multiple architectures
  • Support from the IBM/NVidia Center of Excellence staff at the ORNL as needed
  • Access to early delivery systems and the Summit system as they become available
  • Allocation of compute resources on the full Summit system for the Early Science campaign

 

Proposal submission

Proposals for CAAR partnerships are due February 20, 2015, 23:59 EST, and should be sent electronically to [email protected]. Proposals should address the following elements:

  1. Name and description of the software application, that includes a summary description of the employed methods and algorithms as well as the language(s) and programming model used
  2. Description of the current degree of parallelization of the application and programming model (MPI, MPI+X, PGAS, etc.), available benchmark data illustrating its current scalability, and a description of the anticipated challenges to achieve scalable performance on the Summit architecture
  3. Description of scientific libraries used, especially if the readiness of critical libraries for emerging architectures may be a challenge
  4. Description of the existing software development and distribution practices, including version control and the software release mechanism
  5. Plan for inclusion of code modifications and optimizations into the main branch of the application, including a strategy for the CAAR team to integrate with other development efforts of the application, and a strategy for achieving architecture and performance portability
  6. Summary of the use of the application by the broader scientific and engineering research community, including an estimate of the size of the user base of the application
  7. Summary of the expected impact that a highly optimized version of this application, used on leadership computing resources, will have on the science or engineering field
  8. Vision for a scientific campaign for the Early Science phase of the CAAR project
  9. Composition of the proposing team, including names, titles, level of experience, and levels of effort of the core application developers assigned to carry out the porting and optimization activities during the four-year CAAR project

Proposals are expected to be 12-18 pages in length. Any questions regarding the call for proposals can be addressed to [email protected].

 

Selection criteria

CAAR projects will be selected on the basis of their anticipated impact on the science and engineering fields, the importance to the user programs of the OLCF, the feasibility to achieve scalable performance on Summit, the anticipated opportunity to achieve performance portability for other architectures, the algorithmic and scientific diversity of the suite of CAAR applications. Decisions will be made by the OLCF Scientific Computing staff, in consultation with the IBM/NVidia Center of Excellence at Oak Ridge National Laboratory and the DOE Office of Advanced Scientific Computing Research.

You can attend to information webinars on Dec 10th, 2014 and Jan 6th, 2015. Registration for this webinar can be found here.

Webinar Presentation: Center for Accelerated Application Readiness – Call for Proposals: Getting Applications Ready for Summit