Our Success Stories, a term coined to represent the success of research projects run at the U.S. Army Research Laboratory (ARL) DoD Supercomputing Resource Center (DSRC), continue to represent our role as an HPC organization serving DoD scientists and researchers. The ARL DSRC continues in its pursuit to help provide insight and answers to some of the DoD's toughest problems. Recent ARL DSRC Success Stories include:
Advances In Multiscale Predictions Of Turbulent Spray Formation And Impinging Jet Atomization, To support the Warfighter, the Army needs to provide reliable and efficient propulsion systems for heavy fuel engine platforms that exclusively rely on direct injection fuel delivery systems. Read More.
Spectral Transforms And Eigensolvers For Exascale Computing, New computing architectures present challenges for traditional eigenvalue solvers; in particular, communication becomes increasingly expensive. Our goal is to develop methods of avoiding communication to effect more scalable eigenvalue solver software. Read More.
Programmable Network Algorithms & Models For Classical & Quantum Networks, Software defined programmable network abstractions and control plane intelligence is leading the much awaited network innovation. Programmable networking offers device agnostic communication framework to encode new network functions. Read More.
Stanford And The Army Set Computational Record, Stanford Professor Charbel Farhat and his team accomplished a rare feat in computer engineering through a partnership with the U.S. Army Research Laboratory. Read More.
High Performance Computing For Geospatial, Vehicular, And Network Data Analysis, Growing dependence on digital communications and the availability of highly sophisticated communications systems require more complex and detailed test events, producing terabytes of data per day. Read More.
Auto-Tuning Benchmark For HPC Accelerators, Current high performance computing platforms employ a diverse range of resources to achieve maximum performance: shared and distributed memory, accelerators, many cores, multi-core integrated chips, FPGAs, etc. and we know the future portends even greater heterogeneity. Read More.
Multi-Objective Geometric Optimization For Heterogeneous Architectures, ARL has developed a method for performing multi-objective geometric optimization in the context of optimally positioning blue force Soldiers to reduce the risk posed by ballistic threats while maximizing the observability of surveillance targets identified by mission requirements. Read More.
CFD Simulation Of Maneuvering Projectile Aerodynamics, For maneuvering munitions, the effect of many weapon control mechanisms such as canards, reaction jets, deployable pins, pulsed flaps, and micro-jets on aerodynamics and flight dynamics is critical to overall guided flight performance. Read More.
Atomistic Simulation Of An Explosive Formulation (Composition A3 Type II), In order to enable a computational capability for design of next generation energetic materials (EMs), ARL has made a large investment in the development of a multi-scale modeling and simulation (MM&S) paradigm that incorporates all of the length scales (atomistic through continuum) that have to be addressed in order to computationally characterize an explosive. Read More.
Programming Approaches For Neuromorphic Cognitive Computing, Neuromorphic computing is an emerging technology that promises intelligent real-time computing at extremely low power, less than 1 watt. Our research focuses on discovering new neuromorphic programming techniques, and computing paradigms that will enable the development of future low-power adaptable cognitive computing systems. Read More.