Early Science Program

ALCF Aurora Early Science Program

Aurora Announcement

On 18 March 2019, DOE, ANL, Intel, and Cray announced Aurora will be the nation's first exascale computer.

On the ESP

As part of the process of bringing a new supercomputer into production, the ALCF hosts the Early Science Program (ESP) to ensure its next-generation systems are ready to hit the ground running. The intent of the ESP is to use the critical pre-production time period to prepare key applications for the architecture and scale of a new supercomputer, and to solidify libraries and infrastructure to pave the way for other production applications to run on the system. In addition to fostering application readiness, the ESP allows researchers to pursue innovative computational science projects not possible on today’s leadership-class supercomputers.

The current phase of ESP is now targeting the nation's first exascale system, to be delivered to ALCF in 2021. 

We have already selected a set of 10 projects in the traditional HPC simulation pillar. These projects are underway, working toward Aurora 2021. The call for simulation-based proposals for the Aurora ESP was open from July 6 to September 2, 2016. On January 30, 2016, we announced these Aurora ESP projects.

ALCF's mission is now to officially support not just traditional simulation-based HPC, but to support the three pillars of Simulation, Data, and Learning:

• Simulation - traditional simulation-based computational science
• Data - data-centric and data-intensive computing at leadership scale
• Learning - machine learning, deep learning, and other AI areas revolutionizing scientific HPC

Our most recent call for proposals specifically sought proposals with Data and Learning focus (but was open to proposals that are cross-cutting across the three pillars). We selected 10 new projects from this. 

Aurora Tools and Libraries Project

To facilitate development of community software tools and libraries critical to Aurora, especially those needed by applications in ESP projects, ALCF awarded and manages a special omnibus project: the Aurora Tools and Libraries Project. Scott Parker serves as coordinating PI, with multiple PIs leading individual development projects. This team of teams shares the same early access to information, software, and hardware as the applications ESP projects, and works collaboratively with ALCF and the vendors.

Aurora ESP Postdoctoral Opportunities

Join us in launching the first scientific calculations on our next-generation leadership system, Aurora. We are looking for motivated candidates for two-year ALCF postdoctoral appointments working with many of our projects. You'll find brief descriptions of the positions, and links for applying, on our Careers page.

Previous Iterations of ESP

The previous ESP phase targeted Theta, a system based on Intel’s second-generation Xeon Phi processor, delivered to ALCF in late 2016, Theta will serve as a bridge between the ALCF’s other production system, Mira, and the next leadership-class supercomputer, Aurora, an exascale system slated for delivery in 2021. The Theta ESP call for proposals was open from April 22 - May 22, 2015, and projects were selected in July 2015. For details, see the Theta ESP project announcement.

The Theta and Aurora Early Science Programs are based on the ALCF's highly successful Mira ESP.

ALCF Theta ESP Technical Reports

These reports document the computational efforts and lessons learned in preparing the projects' applications to run on Theta. These Argonne/ALCF reports may form the basis of journal/conference publications:

• Electronic Structure-Based Discovery of Hybrid Photovoltaic Materials on Next-Generation HPC Platforms (ANL/ALCF/ESP-17/2)
• Large-Scale Simulation of Brain Tissue, Blue Brain Project, EPFL (ANL/ALCF/ESP-17/3)
• Flow, Mixing and Combustion of Transient Turbulent Gaseous Jets in Confined Cylindrical Geometries (ANL/ALCF/ESP-17/4)
• First-Principles Simulations of Functional Materials for Energy Conversion (ANL/ALCF/ESP-17/5)
• Advanced Electronic Structure Methods for Heterogeneous Catalysis and Separation of Heavy Metals (ANL/ALCF/ESP-17/6)
• Next-Generation Cosmology Simulations with HACC: Conquering the Baryon Problem (ANL/ALCF/ESP-17/7)
• Extreme Scale Unstructured Adaptive CFD: From Multiphase Flow to Aerodynamic Flow Control (ANL/ALCF/ESP-17/8)
• The Hadronic Contribution to the Anomalous Magnetic Moment of the Muon (ANL/ALCF/ESP-17/9)
• Quantum Monte Carlo Calculations in Nuclear Theory (ANL/ALCF/ESP-17/10)
• Free Energy Landscapes of Membrane Transport Proteins (ANL/ALCF/ESP-17/11)
   

ALCF Theta Early Science Program Presentations at  IXPUG 2016

Multiple projects from the ALCF Theta Early Science Program gave presentations at the 2016 IXPUG Annual U.S. Meeting on preparing applications for Theta:

• Williams, "ALCF Early Science Program for Xeon Phi Supercomputers"
• Li et al., "MILC Staggered Conjugate Gradient Performance on Intel KNL"
• Boyle et al., "Grid: Structured Cartesian Mesh Library for Quantum Chromodynamics"
• Finkel et al., "HACC on the KNL - Porting, Optimizing, and Early Experiences"
• Joo et al., "Chroma on Knights Landing Three Ways"

Several other talks related indirectly, with ALCF contribution:

• Luo et al., "Optimizations of Bspline-based Orbital Evaluations in Quantum Monte Carlo on Multi/Many-core Shared Memory Processors"
• DiStasio et al., "Enabling Large-Scale Hybrid Density Functional Theory Based Ab Initio Molecular Dynamics in Condensed-Phase Systems"
• Alexeev et al., "Efficient MPI/OpenMP Parallelization of the Hartree-Fock Method on Intel® Xeon Phi Processors"
• Finkel, "LLVM and Clang on the KNL"
• Rizzi et al., "Visualization with vl3. Now on the CPU"