This message serves as official notice that the DOW High Performance Computing Modernization Program (HPCMP) is soliciting interest in submission of proposals from the HPCMP user community addressing a need for HPC resources that cannot be obtained through usage of standard HPC resources available in any of the five DSRCs.
The first step in this process requires interested parties to completely fill-out the attached form entitled “Indication of Interest in Submitting a Dedicated HPC Initiatives (DHPI) Proposal to HPCMP” (Click Here). All submissions will be reviewed and ranked by the appropriate Service/Agency Principals for HPCMP. A small subset of these entries will be selected to proceed to full proposal that provides greater detail into the proposed work, its impact to DOW, and a justification for DHPI resources. Please be advised that the cap for the DHPI hardware funding is set at $2M/system without written consent from HPCMP Leadership. DHPI systems must be co-located at one of the HPCMP DSRCs unless written approval is obtained from HPCMP Leadership for alternate location.
It is important to note that the DHPI funding only covers purchase of the hardware—all other costs incurred including labor, commercial or licensed software, contracting services, facility preparation, specialized networking solutions, extended warranties, and maintenance shall be covered by the proposing organization.
Please address all questions to Dr. Frances Hill (
Please submit your completed form indicating interest in the DHPI process to Ms. Crystal Chancio (
Sage is an integrated software tool for training surrogate models, assessing surrogate model quality, and refining surrogate models through intelligent adaptive sampling processes. Sage provides the following key features:
- Data Fitting Algorithms: The framework includes algorithms for fitting available data to form surrogate models. It ensures that the model captures essential features while remaining computationally tractable.
- Sampling Techniques: The framework provides techniques for intelligently selecting samples over the input space. Key adaptive sampling techniques leverage surrogate models to focus samples where they most improve model accuracy.
- Model Validation: The framework provides techniques for assessing the accuracy of the model.
- Uncertainty Quantification: The trained models provide estimates of their confidence in their predictions.
Applicability of CREATE Tools and How to Apply for a CREATE Account
Any government employee, contractor, or academic researcher who is actively supporting the DoD's RDT&E and/or acquisition engineering mission may request access to the HPCMP CREATE™ software. Access to this software does not depend on being an active HPCMP user.
Click here to apply for a CREATE License.
Once approved, visit https://create.hpc.mil to access CREATE Resources, for information on how to apply for a Portal to the Information Environment (pIE) Account, or for Technical Support Inquiries.
After applying, if you have any questions about your application status or access, please contact CREATE User Accounts at
Questions for CREATE
For questions about anything CREATE related, contact our
Technical Support Inquiries
For technical support with CREATE products, contact the appropriate software product by selecting it below:
HPCMP Account Creation
To obtain an account on HPCMP supercomputing resources, please visit Obtaining an Account and follow the instructions for obtaining an HPC systems account.
The overarching goals of the CREATE-Ships project are to develop software that enables comprehensive exploration of trade-space design options for complex maritime systems; provide confidence in the results of computational predictions in required disciplines across all phases of the acquisition process; and meet required acquisition timelines for DoD clients, primarily the US Navy. Our software suite includes: RSDE, IHDE, NESM, and ISDE.
RSDE is a concept design tool that allows engineers and naval architects to assess the trade-offs inherent in designing ships to meet a spectrum of competing performance parameters. RSDE provides the following critical capabilities: 1) Generates a wide range of ship concept options with corresponding design and analysis data, 2) Supports a set-based design down-selection process leading to the discovery of low-cost, robust, candidate design solutions; and 3) Develops data for physics-based response surfaces to aid in the visualization of design, cost, and performance trade-offs of a multi-dimensional design space.
Employing the concept of design space exploration, RSDE enables engineers and naval architects to provide data for decision makers on the impact of trade-offs in capability, such as range, speed, armament, and aviation support versus cost drivers of a proposed ship concept. RSDE can generate tens of thousands of candidate ship designs with varying hull form size and shape, systems, structures, powering, and payloads. To date, RSDE has been used to support set-based design on numerous Navy acquisition programs, providing data necessary to enable the set-based down-selection process, as well as additional design information and reduced uncertainty. The RSDE effort is also supporting the development of a submarine design space exploration tool to support similar capabilities for submarine platform design.
IHDE is a workbench-like desktop application that integrates a suite of hull form design and analysis tools for users to evaluate hydrodynamic performance, including visualization, in a simplified and timely manner. IHDE provides a highly-automated environment that allows naval architects and hydrodynamicists to build computational models, submit "jobs" to local or HPC resources, and retrieve and validate results for various required areas of hydrodynamics, and at varying levels of fidelity. Ship hull designers are able to assess ship performance in areas of resistance, seakeeping, hydrodynamic loads, and operability in the form of percent time operable (PTO) for several different mission types. The development plan calls for additional capabilities related to ship powering and maneuvering performance characterization and multi-objective hull form optimization.
The IHDE also includes an analysis tool validation engine, which provides the user with validation information by leveraging historical model test data and best-practice precomputed solutions for user-driven comparisons. The advantages of the IHDE include automated analysis preparation, automated grid generation, and integrated visualization. The Leading-Edge Architecture for Prototyping Systems (LEAPS) product model is used to manage geometry and analysis information, and allows data sharing between several knowledge domains.
LEAPS provides a warehouse for ship model information, which provides distinct advantages in maintaining an accurate ship ontology:
1. Interoperability for different activities which already use LEAPS;
2. Creation of a foundation of common terms across different disciplines for ship geometries and characteristics, thereby preventing translation errors; and
3. Synergy in software development and integration using a common product model that can be shared with other activities, eliminating inefficient and time-consuming resetting of the same ship modeling inputs when using different analysis methods.
The IHDE utilizes a LEAPS database as a starting point for any hydrodynamic analysis, and is further enabled by the use of the LEAPS Morpheus preprocessor to define the ship model from CAD information. The end-state vision for the IHDE is to provide an integrated suite of design and analysis tools that cover a range of fidelity and corresponding computational expense, in order to fully characterize a ship design concept across relevant ship performance areas with an appropriate level of definition. The advantages of the IHDE include automation of analysis preparation, improving efficiency and reducing input errors, automated grid generation, parallel execution, and integrated visualization capabilities.
NESM builds on the Department of Energy Sandia National Laboratory's multi-physics toolkit, Sierra Mechanics, to provide a means to assess ship and component response to external shock and blast using accurate HPC tools. When used within a comprehensive shock program, NESM can reduce the time and expense associated with physical shock testing of ship classes by leveraging physics-based modeling capabilities. It also supports a process to improve initial ship design by assessing planned component installations for shock resilience prior to final arrangement and installation decisions. The coupled multi-physics capabilities include structural dynamics (implicit linear-elastic solvers), solid mechanics (explicit nonlinear solvers), fluid dynamics (Euler solvers), and fluid-structure interaction.
The solution algorithms in NESM exploit massively parallel computers and scale to tens-of-thousands of CPUs, enabling efficient computer use and the ability to address full-sized naval vessels up to and including next-generation aircraft carriers and submarines.
NESM is being used to support Live-Fire Test and Evaluation requirements for the new Ford Class nuclear powered aircraft carriers. All ship classes are required to be tested for shock damage resistance. Historically this has been done exclusively via physical testing using underwater explosions. Effects were recorded and necessary alterations were designed and implemented for the ship class. This not only required a ship to be dedicated to the lengthy test process, but also requires any deficiencies in shock resilience to be retrofitted into the ship design. NESM enables an improved shock design process including computational assessment of equipment survivability while still in the design stage, when fixes can be incorporated prior to construction. NESM has been approved by the Navy as the M&S tool to achieve the shock trial alternative process.
The Integrated Structural Design Environment (ISDE) is developing the framework for a suite of structural design and analysis tools ranging from low-fidelity concept design to high-fidelity structural analysis tools. It will consist of a suite of structural design, analysis, and visualization capabilities that will inform and improve structural design decisions, and provide rapid response for in-service ship incidents. ISDE will integrate with the CREATE-Ships Rapid Ship Design Environment (RSDE) through the Leading Edge Architecture for Prototyping Systems (LEAPS) product model. The integration will expand the use of RSDE beyond just early-stage concept design by supporting a range of fidelity and improved structural weight estimates. ISDE will enable multi-level design optimization (MDO) to reduce cost, improve structure, and allow for longer service life. Use of ISDE in the HPC environment will allow for high-fidelity, rapid structural analysis response to ship incidents during service life. ISDE will also connect to the CREATE-Ships NESM tool in order to perform high-fidelity analyses.
The objective of the ISDE suite of tools as part of the CREATE-Ships program is to develop and automate the ship and submarine structural design capabilities and analysis tools needed to meet current and future navy acquisition programs, and to support in-service structural engineering efforts.
The HPCMP CREATE™-SH ISDE tool will be a multi-fidelity structural analysis environment to address the inadequacies in current structural design tools by providing a computation data-bridge between structural concept design, first-order assessments, and detailed FEA while introducing the capabilities needed for future Navy requirements. ISDE will provide a technically proficient structural design environment built on robust software development processes. Modular structural design/analysis will ensure inclusion of current and future structural capabilities. ISDE will include comparison of fatigue life, life cycle costs, SOE and maintenance planning. The software tool will also enable integrated incident response room evaluations.
As integral components of almost all DoD weapon systems, antennas and radars perform critical communication, identification, and navigation functions necessary for warfighter effectiveness and survivability. Computational Electromagnetics (CEM) tools have matured in recent years and are widely used for antenna concept exploration and design by the DoD and its contractors. Our software suite includes: SENTRi Generation 1 (G1), SENTRi Generation 2 (G2), and Aurora.
SENTRi is CREATE-RF's flagship product. It is a full wave computational electromagnetic (CEM) code utilizing the hybrid finite element boundary integral method (FEBI), making it suitable for solving a wide variety of electromagnetic problems including antennas, antennas on platforms, passive microwave circuits, and EMI/EMC. SENTRi is available for both MS Windows and Linux operating systems. It can be run on engineering workstations and utilizes MPI technology allowing it to take full advantage of the HPCMP DSRC supercomputing systems. SENTRi comes in two versions. SENTRi Generation 1 (G1) is a highly optimized solver for monolithic meshes with several advanced capabilities. SENTRi Generation 2 (G2) was developed due to technical limitations inherent in the Generation 1 code. Additional features in SENTRi G2 include a richer BI solver, the domain decomposition method (DDM), hierarchical basis functions, symmetry, and a powerful Python scripting interface.
Frequency Domain
SENTRi is a frequency domain code — which means one frequency at a time is solved; frequency sweeps are handled by running a series of solutions as defined by the user. SENTRi was designed as a frequency domain code to provide a higher degree of accuracy. In addition, frequency domain methods provide faster solutions for far field sweeps, calculation of S-parameter matrices for multi-port microwave circuits, and scanning of phased array antennas.
Material Modeling
SENTRi is designed for the modeling of complex structures including highly, heterogeneous material structures with multi-scaled features. SENTRi's material modeling includes dielectric, magnetic, impedance boundary conditions, and complex valued resistive sheets. Volume materials can have combined dielectric and magnetic properties, and can be isotropic or anisotropic. Volume materials can also have continuously varying properties with definitions provided by user-authored Python scripts. Similarly, Python scripts are used for defining the frequency varying properties of dispersive materials and for position varying resistive sheets.
Hybridized with SENTRi G1 and available from the CREATE-RF program, Aurora is an asymptotic approximate EM solver, employing the physical-optics shooting-and-bouncing-rays (PO-SBR) technique for computing installed antenna patterns and plane wave scattering from electrically large platforms. Helping fulfill CREATE-RF's antenna-on-platform integration analysis objective, Aurora has modeled antenna patterns from parabolic reflectors up to aircraft carriers. It supports Windows and Linux and executes efficiently on HPC systems.
Hybrid SENTRi/Aurora In-Situ Antenna Analysis
SENTRi accurately predicts antenna patterns. As frequency rises, modeling platform interaction grows to be intractable. The SENTRi/Aurora hybrid accelerates and makes tractable platform integration analysis on otherwise unreachable problems. Reduced solve times also aid engineers in quickly understanding the impact of features such as propellers on the CV-22 Osprey at 2.4 GHz in less than an hour on a laptop.
Large Scale Problem Capable
Using 8,192 cores on the Excalibur system at the ARL DSRC, Aurora simulated a 2.4-GHz horizontal dipole antenna on a Nimitz class carrier with 13 FA-18 Hornets in 7 minutes.