U.S. Army plans for a 100 petaflop supercomputer

HPC will play a vital role in future Army strategy planning.

The US Army Research Laboratory (ARL) has laid out its needs to maintain competitive and tactical advantages for its land forces, and high-performance computing (HPC) is at the heart of it. In a newly released plan covering the next four years, the Army sees advanced computing as something central to its mission.

In Technical Implementation Plan for 2015-2019 (PDF format), the Army acknowledges that budgets will be reduced in the coming years, but still lays out a plan from this year to 2030, with technology as one of seven major areas of focus. The other areas are materials research, sciences-for-maneuver, information sciences, sciences-for-lethality and protection, human sciences, and assessment and analysis.

The Computational Sciences Campaign consists of three initiatives: Tactical High Performance Computing (HPC) to maintain the superiority of Army material systems through predictive modeling and simulation technologies; Very Large-scale Data Analytics for the Army to facilitate information dominance, distributed maneuver operations, and human sciences through computational data intensive sciences; and Computational Predictive Design for Interdisciplinary Sciences to significantly increase and tailor advanced computing architectures and computing sciences technologies on the forefront to enable land power dominance.

Tactical High Performance Computing, the first part of the plan, lists a 100 petaflop supercomputer as central to improving mission effectiveness. The system would enable real-time processing for Soldiers operating at the tactical edge and improve mission effectiveness and mitigate risk in hostile environments. Such a computer would be twice the power of the current top supercomputer in the world, China's Tianhe-2, which clocks in at 54 petaflops.

"Achieving such a system with current computing devices which are constrained devices by power and performance is untenable. By aggregating the computing processing power of deployed friendly computing devices through distributed computing, supplemented by the projection of mobile customized HPC platforms operating at the tactical edge (tactical cloudlets), a new level of capabilities is possible for mounted and dismounted Soldiers," the Army said in its proposal.

The second core program is very large-scale data analytics. The Army is clearly looking at a Big Data program with this despite not using the term, as it plans to mine its data stores, generated from battlefield networks, sensors, experiments, observations, and numerical simulations, to provide information to commanders across a range of military operations.

"The ARL sees extraordinary potential for scientific advance inherent in large-scale complex data, but these benefits hinge on the development of scalable computational methods and massively parallel hierarchical computing architectures," it wrote.

Computational predictive design for interdisciplinary sciences will shorten the development cycle and substantially improves performance of lethality, protection, electronics, power, and dismounted soldier gear utilizing lightweight, multi-functional, cost effective, optimized innovative materials by design exploiting high performance computers. In plain English, they are looking to build better, lightweight combat uniforms.

"The materials engineered through predictive design computational methods can be fabricated according to Army specification and with optimized performance at every spatial and temporal scale. Materials subjected to extreme conditions such as mechanical shock, pressure and electromagnetic fields are of particular importance to the Army and require advanced multi-scale and multi-physics computational strategies for successful engineering design," Army researchers wrote.

Will we taxpayers ever see these technological advances? Possibly. Don't forget that the Internet you are now using started as a military program to create a redundant network designed to survive a nuclear war.

What’s wrong? The new clean desk test
Join the discussion
Be the first to comment on this article. Our Commenting Policies