Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory Optimization”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
NASA NTRS1992

The design and implementation of a parallel unstructured Euler solver using software primitives

This paper is concerned with the implementation of a three-dimensional unstructured grid Euler-solver on massively parallel distributed-memory computer architectures. The goal is to minimize solution time by achieving high computational rates with a numerically efficient algorithm. An unstructured multigrid algorithm with an edge-based data structure has been adopted, and a number of optimizations have been devised and implemented in order to accelerate the parallel communication rates. The implementation is carried out by creating a set of software tools, which provide an interface between the parallelization issues and the sequential code, while providing a basis for future automatic run-time compilation support. Large practical unstructured grid problems are solved on the Intel iPSC/860 hypercube and Intel Touchstone Delta machine. The quantitative effect of the various optimizations are demonstrated, and we show that the combined effect of these optimizations leads to roughly a factor of three performance improvement. The overall solution efficiency is compared with that obtained on the CRAY-YMP vector supercomputer.

Das, R.↗
NASA NTRS1994

Design and implementation of a parallel unstructured Euler solver using software primitives

This paper is concerned with the implementation of a three-dimensional unstructured-grid Euler solver on massively parallel distributed-memory computer architectures. The goal is to minimize solution time by achieving high computational rates with a numerically efficient algorithm. An unstructured multigrid algorithm with an edge-based data structure has been adopted, and a number of optimizations have been devised and implemented to accelerate the parallel computational rates. The implementation is carried out by creating a set of software tools, which provide an interface between the parallelization issues and the sequential code, while providing a basis for future automatic run-time compilation support. Large practical unstructured grid problems are solved on the Intel iPSC/860 hypercube and Intel Touchstone Delta machine. The quantitative effects of the various optimizations are demonstrated, and we show that the combined effect of these optimizations leads to roughly a factor of 3 performance improvement. The overall solution efficiency is compared with that obtained on the Cray Y-MP vector supercomputer.

Das, R.↗
NASA NTRS2004

An Integrated Decision-Making Model for Categorizing Weather Products and Decision Aids

The National Airspace System s capacity will experience considerable growth in the next few decades. Weather adversely affects safe air travel. The FAA and NASA are working to develop new technologies that display weather information to support situation awareness and optimize pilot decision-making in avoiding hazardous weather. Understanding situation awareness and naturalistic decision-making is an important step in achieving this goal. Information representation and situation time stress greatly influence attentional resource allocation and working memory capacity, potentially obstructing accurate situation awareness assessments. Three naturalistic decision-making theories were integrated to provide an understanding of the levels of decision making incorporated in three operational situations and two conditions. The task characteristics associated with each phase of flight govern the level of situation awareness attained and the decision making processes utilized. Weather product s attributes and situation task characteristics combine to classify weather products according to the decision-making processes best supported. In addition, a graphical interface is described that affords intuitive selection of the appropriate weather product relative to the pilot s current flight situation.

Elgin, Peter D.↗
DOE OSTI2024

Distributed Multi-GPU Community Detection on Exascale Computing Platforms

Community detection is a fundamental operation in graph mining, and by uncovering hidden structures and patterns within complex systems it helps solve fundamental problems pertaining to social networks, such as information diffusion, epidemics, and recommender systems. Scaling graph algorithms for massive networks becomes challenging on modern distributed-memory multi-GPU (Graphics Processing Unit) systems due to limitations such as irregular memory access patterns, load imbalances, higher communication-computation ratios, and cross-platform support. We present a novel algorithm HiPDPL-GPU (distributed parallel Louvain) to address these challenges. We conduct experiments involving different partitioning techniques to achieve optimized performance of HiPDPL-GPU on the two largest supercomputers: Frontier and Summit. Remarkably, HiPDPL-GPU processes a graph with 4.2 billion edges in less than 3 minutes using 1024 GPUs. Qualitatively performance of HiPDPL-GPU is similar or better compared to other state-of-the-art CPU- and GPU-based implementations. While prior GPU implementations have predominantly employed CUDA, our first-of-its-kind implementation for community detection is cross-platform, accommodating both AMD and NVIDIA GPUs.

graph algorithms, high performance comptuing↗
DOE OSTI2026

Xanthos-Lake Model Source Code

This repository contains the source code for Xanthos-Lake, a lake-modeling extension of the Xanthos framework that introduces a coupled lake component comprising the Xanthos-Lake Snow and Ice Model (xLSIM) and the Xanthos-Lake Water Balance Model (xLWBM). xLSIM is a basin-aware machine-learning model for lake snow, ice, and thermal conditions. It predicts monthly lake ice thickness, snow depth, snow-cover fraction, mixing-layer temperature, and lake ice fraction from meteorological forcing and lake surface-area information. It uses sequence-based deep-learning architectures, including Transformer and hybrid Long Short-Term Memory–Transformer (LSTM–Transformer) models, together with seasonal encoding, multi-lake learning, physical masking, and basin-level cryospheric and non-cryospheric classification. The training workflow uses Ray for scalable execution and includes optional Ray Tune hyperparameter optimization. Model predictions, observations, diagnostics, and feature-importance outputs are written in NetCDF. xLWBM is the water-balance component of the new lake framework. It simulates monthly lake storage, surface area, evaporation, inflow, outflow, and lake–groundwater exchange. It combines physical water-balance equations with calibrated bathymetric relationships, weir-based outlet flow, modified Penman open-water evaporation, groundwater head relaxation, Penman–Monteith snow and ice sublimation, and snow, ice, and thermal conditions supplied by xLSIM. The model calibrates lake parameters against satellite-derived surface-area data, using evaporation-based calibration where surface-area data are unavailable, and supports small, medium, and large lake classes. For large lakes, xLWBM is integrated with the managed-routing workflow so that lake storage and outflow interact directly with downstream river routing and reservoir operations. Together, xLSIM and xLWBM provide Xanthos with a coupled lake-modeling capability. xLSIM supplies the snow, ice, and thermal conditions that affect lake evaporation and snow- and ice-related water exchanges, while xLWBM translates those conditions into dynamic lake storage, surface area, evaporation, and discharge. In return, xLWBM supplies evolving lake surface area to xLSIM. This coupling enables Xanthos to represent lakes as active hydrologic components within basin-scale water-availability and routing simulations.

Machine Learning↗
NASA NTRS1991

A parallel algorithm for multi-level logic synthesis using the transduction method

The Transduction Method has been shown to be a powerful tool in the optimization of multilevel networks. Many tools such as the SYLON synthesis system (X90), (CM89), (LM90) have been developed based on this method. A parallel implementation is presented of SYLON-XTRANS (XM89) on an eight processor Encore Multimax shared memory multiprocessor. It minimizes multilevel networks consisting of simple gates through parallel pruning, gate substitution, gate merging, generalized gate substitution, and gate input reduction. This implementation, called Parallel TRANSduction (PTRANS), also uses partitioning to break large circuits up and performs inter- and intra-partition dynamic load balancing. With this, good speedups and high processor efficiencies are achievable without sacrificing the resulting circuit quality.

Lim, Chieng-Fai↗
NASA NTRS1992

The design and implementation of a parallel unstructured Euler solver using software primitives

This paper is concerned with the implementation of a 3D unstructured-grid Euler-solver on massively parallel distributed-memory computer architectures. The goal is to minimize solution time by achieving high computational rates with a numerically efficient algorithm. An unstructured multigrid algorithm with an edge-based data-structure has been adopted, and a number of optimizations have been devised and implemented in order to accelerate the parallel computational rates. The implementation is carried out by creating a set of software tools, which ease the implementation of computational problems on parallel architecture machines by relieving the user of the low-level machine specific issues. The quantitative effect of the various optimizations are demonstrated, and we show that the combined effect of these optimizations leads to roughly a factor of three performance improvement. The overall solution efficiency is compared with that obtained on the CRAY-YMP vector supercomputer.

Das, R.↗
NASA NTRS2005

An Overview of Wind-Driven Rovers for Planetary Exploration

The use of in-situ propulsion is considered enabling technology for long duration planetary surface missions. Most studies have focused on stored energy from chemicals extracted from the soil or the use of soil chemicals to produce photovoltaic arrays. An older form of in-situ propulsion is the use of wind power. Recent studies have shown potential for wind driven craft for exploration of Mars, Titan and Venus. The power of the wind, used for centuries to power wind mills and sailing ships, is now being applied to modern land craft. Efforts are now underway to use the wind to push exploration vehicles on other planets and moons in extended survey missions. Tumbleweed rovers are emerging as a new type of wind-driven science platform concept. Recent investigations by the National Aeronautics and Space Administration (NASA) and Jet Propulsion Laboratory (JPL) indicate that these light-weight, mostly spherical or quasi-spherical devices have potential for long distance surface exploration missions. As a power boat has unique capabilities, but relies on stored energy (fuel) to move the vessel, the Tumbleweed, like the sailing ships of the early explorers on earth, uses an unlimited resource the wind to move around the surface of Mars. This has the potential to reduce the major mass drivers of robotic rovers as well as the power generation and storage systems. Jacques Blamont of JPL and the University of Paris conceived the first documented Mars wind-blown ball in 1977, shortly after the Viking landers discovered that Mars has a thin CO2 atmosphere with relatively strong winds. In 1995, Jack Jones, et al, of JPL conceived of a large wind-blown inflated ball for Mars that could also be driven and steered by means of a motorized mass hanging beneath the rolling axis of the ball. A team at NASA Langley Research Center started a biomimetic Tumbleweed design study in 1998. Wind tunnel and CFD analysis were applied to a variety of concepts to optimize the aerodynamic characteristics of the Tumbleweed Rovers. Bare structures, structures carrying sails and a tumbleweed plant (of the Salsola genus) were tested in Langley's wind tunnels. Thomas Estier of the Swiss Federal Institute of Technology developed a memory metal collapsible structure, the Windball. Numerous other researchers have also suggested spherical rovers.

Hajos, Gregory A.↗
NASA NTRS2013

A Mars Sample Return Sample Handling System

We present a sample handling system, a subsystem of the proposed Dragon landed Mars Sample Return (MSR) mission [1], that can return to Earth orbit a significant mass of frozen Mars samples potentially consisting of: rock cores, subsurface drilled rock and ice cuttings, pebble sized rocks, and soil scoops. The sample collection, storage, retrieval and packaging assumptions and concepts in this study are applicable for the NASA's MPPG MSR mission architecture options [2]. Our study assumes a predecessor rover mission collects samples for return to Earth to address questions on: past life, climate change, water history, age dating, understanding Mars interior evolution [3], and, human safety and in-situ resource utilization. Hence the rover will have "integrated priorities for rock sampling" [3] that cover collection of subaqueous or hydrothermal sediments, low-temperature fluidaltered rocks, unaltered igneous rocks, regolith and atmosphere samples. Samples could include: drilled rock cores, alluvial and fluvial deposits, subsurface ice and soils, clays, sulfates, salts including perchlorates, aeolian deposits, and concretions. Thus samples will have a broad range of bulk densities, and require for Earth based analysis where practical: in-situ characterization, management of degradation such as perchlorate deliquescence and volatile release, and contamination management. We propose to adopt a sample container with a set of cups each with a sample from a specific location. We considered two sample cups sizes: (1) a small cup sized for samples matching those submitted to in-situ characterization instruments, and, (2) a larger cup for 100 mm rock cores [4] and pebble sized rocks, thus providing diverse samples and optimizing the MSR sample mass payload fraction for a given payload volume. We minimize sample degradation by keeping them frozen in the MSR payload sample canister using Peltier chip cooling. The cups are sealed by interference fitted heat activated memory alloy caps [5] if the heating does not affect the sample, or by crimping caps similar to bottle capping. We prefer cap sealing surfaces be external to the cup rim to prevent sample dust inside the cups interfering with sealing, or, contamination of the sample by Teflon seal elements (if adopted). Finally the sample collection rover, or a Fetch rover, selects cups with best choice samples and loads them into a sample tray, before delivering it to the Earth Return Vehicle (ERV) in the MSR Dragon capsule as described in [1] (Fig 1). This ensures best use of the MSR payload mass allowance. A 3 meter long jointed robot arm is extended from the Dragon capsule's crew hatch, retrieves the sample tray and inserts it into the sample canister payload located on the ERV stage. The robot arm has capacity to obtain grab samples in the event of a rover failure. The sample canister has a robot arm capture casting to enable capture by crewed or robot spacecraft when it returns to Earth orbit

Wilson, David↗
DOE OSTI2024

DyG-DPCD: A Distributed Parallel Community Detection Algorithm for Large-Scale Dynamic Graphs

Dynamic (Temporal) graphs capture the valuable evolution of real-world systems, from the continuously evolving patterns of social interactions and genetic pathways to the dynamic fluctuations of economic forces. Detecting communities for such evolving networks poses unique challenges. Detecting and analyzing the evolution of communities within dynamic graphs unlocks valuable insights into the underlying structural and temporal patterns of real-world systems. However, the sheer volume of modern graph data and the inherent complexity of the temporal dimension pose significant challenges to scalable community detection algorithms. Addressing this gap, our work explores the limited landscape of scalable distributed-memory parallel methods specifically designed for dynamic network community detection. We propose a novel parallel algorithm, DyG-DPCD (Dynamic Graph Distributed Parallel Community Detection), to detect communities in dynamic networks using the Message Passing Interface (MPI) framework. We present a vertex-centric approach, allowing us to detect communities through local optimization. Furthermore, we enhance our baseline algorithm by incorporating three heuristics, which improve the algorithm’s performance significantly while maintaining the quality of the solutions. We demonstrate the efficiency of our algorithm by experimenting on several real-world large-scale networks with hundreds of millions of edges spanning diverse domains. Notably, DyG-DPCD achieves speedups between 25× and 30× for large networks that we experimented on using NERSC compute nodes. In conclusion, our algorithm outperforms the STINGER parallel re-agglomeration algorithm by 30×.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1995

GridTool: A surface modeling and grid generation tool

GridTool is designed around the concept that the surface grids are generated on a set of bi-linear patches. This type of grid generation is quite easy to implement, and it avoids the problems associated with complex CAD surface representations and associated surface parameterizations. However, the resulting surface grids are close to but not on the original CAD surfaces. This problem can be alleviated by projecting the resulting surface grids onto the original CAD surfaces. GridTool is designed primary for unstructured grid generation systems. Currently, GridTool supports VGRID and FELISA systems, and it can be easily extended to support other unstructured grid generation systems. The data in GridTool is stored parametrically so that once the problem is set up, one can modify the surfaces and the entire set of points, curves and patches will be updated automatically. This is very useful in a multidisciplinary design and optimization process. GridTool is written entirely in ANSI 'C', the interface is based on the FORMS library, and the graphics is based on the GL library. The code has been tested successfully on IRIS workstations running IRIX4.0 and above. The memory is allocated dynamically, therefore, memory size will depend on the complexity of geometry/grid. GridTool data structure is based on a link-list structure which allows the required memory to expand and contract dynamically according to the user's data size and action. Data structure contains several types of objects such as points, curves, patches, sources and surfaces. At any given time, there is always an active object which is drawn in magenta, or in their highlighted colors as defined by the resource file which will be discussed later.

Samareh-Abolhassani, Jamshid↗
NASA NTRS2008

NASA Tech Briefs, August 2008

Customizable Digital Receivers for Radar Two-Camera Acquisition and Tracking of a Flying Target Visual Data Analysis for Satellites A Data Type for Efficient Representation of Other Data Types Hand-Held Ultrasonic Instrument for Reading Matrix Symbols Broadband Microstrip-to-Coplanar Strip Double-Y Balun A Topographical Lidar System for Terrain-Relative Navigation Programmable Low-Voltage Circuit Breaker and Tester Electronic Switch Arrays for Managing Microbattery Arrays Topics covered include: Lower-Dark-Current, Higher-Blue-Response CMOS Imagers; Fabricating Large-Area Sheets of Single-Layer Graphene by CVD; Support for Diagnosis of Custom Computer Hardware; Providing Goal-Based Autonomy for Commanding a Spacecraft; Dynamic Method for Identifying Collected Sample Mass; Optimal Planning and Problem-Solving; Attitude-Control Algorithm for Minimizing Maneuver Execution Errors; Grants Document-Generation System; Heat-Storage Modules Containing LiNO3 3H2O and Graphite Foam; Precipitation-Strengthened, High-Temperature, High-Force Shape Memory Alloys; Improved Relief Valve Would Be Less Susceptible to Failure; Safety Modification of Cam-and-Groove Hose Coupling; Using Composite Materials in a Cryogenic Pump; Using Electronic Noses to Detect Tumors During Neurosurgery; Producing Newborn Synchronous Mammalian Cells; Smaller, Lower-Power Fast-Neutron Scintillation Detectors; Rotationally Vibrating Electric-Field Mill; Estimating Hardness from the USDC Tool-Bit Temperature Rise; Particle-Charge Spectrometer; Automated Production of Movies on a Cluster of Computers; FIDO-Class Development Rover; and Tone-Based Command of Deep Space Probes Using Ground Antennas.

Source record↗
NASA NTRS2020

Machine Learning Algorithm Performance on the Lucata Computer

A new parallel computing paradigm (processor in memory, or PIM) has recently become available, one that uses many lightweight threads, and where each thread migrates automatically to the memory used by that thread. Our effort focuses on understanding how suitable this architecture is for our application, and whether the hardware can sustain speedups as high as the system size permits. In particular we explore the kind of code optimizations needed, and how well optimized code scales. This paper describes some of the those optimizations, and the payoff in terms of scaling.

Kogge, Peter↗
DOE OSTI2024

Distributed Multi-GPU Community Detection on Exascale Computing Platforms

Community detection is a fundamental operation in graph mining, and by uncovering hidden structures and patterns within complex systems it helps solve fundamental problems pertaining to social networks, such as information diffusion, epidemics, and recommender systems. Scaling graph algorithms for massive networks becomes challenging on modern distributed-memory multi-GPU (Graphics Processing Unit) systems due to limitations such as irregular memory access patterns, load imbalances, higher communication-computation ratios, and cross-platform support. We present a novel algorithm HiPDPL-GPU (Distributed Parallel Louvain) to address these challenges. We conduct experiments involving different partitioning techniques to achieve an optimized performance of HiPDPL-GPU on the two largest supercomputers: Frontier and Summit. Remarkably, HiPDPL-GPU processes a graph with 4.2 billion edges in less than 3 minutes using 1024 GPUs. Qualitatively, the performance of HiPDPL-GPU is similar or better compared to other state-of-the-art CPU- and GPU-based implementations. While prior GPU implementations have predominantly employed CUDA, our first-of-its-kind implementation for community detection is cross-platform, accommodating both AMD and NVIDIA GPUs.

Sattar, Naw Safrin↗
NASA NTRS2017

Shape Memory Alloys (SMAs) for Non-Pneumatic Tires

Conventional mobility elements, such as pneumatic tires, suffer from a number of issues related to reliability. Two of the more prevalent problems are the high likelihood of single point failure owing to puncture (i.e. flat tire), and loss of efficiency due to reduction in tire pressure over time. In order to overcome these limitations, alternative compliant tire designs not requiring pneumatics have been developed. However, although current designs have significantly reduced the aforementioned issues, they tend to have their own set of limitations. First, non-pneumatic tires designed for high load applications often have restricted envelopment capability, making their performance less than optimal, especially on uneven terrain. Second, tires designed with larger envelopment capability tend to suffer from large amounts of plasticity (permanent deformation) or failure (rupture). Both of these limitations are the direct result of the choice of material being used for the design; conventional metals undergo plastic deformation at low strain while elastomer based designs are often too rigid for the localized deformations needed for high envelopment. Recent advancements at the NASA Glenn research center in a unique class of metals know as shape memory alloys (SMAs) has opened the design space for non-pneumatic compliant tire technologies allowing designs to incorporate orders of magnitude more deformation without damage. The work presented herein highlights the advantages of using SMAs as compared to conventional metals. Additionally, the development of a unique SMA compliant tire design capable of carrying up to 8.9 kN (2000 lbf) with reversible, local deformations on the order of the side wall height will be presented.

Padula, Santo↗
NASA NTRS2000

Incremental Parallelization of Non-Data-Parallel Programs Using the Charon Message-Passing Library

Message passing is among the most popular techniques for parallelizing scientific programs on distributed-memory architectures. The reasons for its success are wide availability (MPI), efficiency, and full tuning control provided to the programmer. A major drawback, however, is that incremental parallelization, as offered by compiler directives, is not generally possible, because all data structures have to be changed throughout the program simultaneously. Charon remedies this situation through mappings between distributed and non-distributed data. It allows breaking up the parallelization into small steps, guaranteeing correctness at every stage. Several tools are available to help convert legacy codes into high-performance message-passing programs. They usually target data-parallel applications, whose loops carrying most of the work can be distributed among all processors without much dependency analysis. Others do a full dependency analysis and then convert the code virtually automatically. Even more toolkits are available that aid construction from scratch of message passing programs. None, however, allows piecemeal translation of codes with complex data dependencies (i.e. non-data-parallel programs) into message passing codes. The Charon library (available in both C and Fortran) provides incremental parallelization capabilities by linking legacy code arrays with distributed arrays. During the conversion process, non-distributed and distributed arrays exist side by side, and simple mapping functions allow the programmer to switch between the two in any location in the program. Charon also provides wrapper functions that leave the structure of the legacy code intact, but that allow execution on truly distributed data. Finally, the library provides a rich set of communication functions that support virtually all patterns of remote data demands in realistic structured grid scientific programs, including transposition, nearest-neighbor communication, pipelining, gather/scatter, and redistribution. At the end of the conversion process most intermediate Charon function calls will have been removed, the non-distributed arrays will have been deleted, and virtually the only remaining Charon functions calls are the high-level, highly optimized communications. Distribution of the data is under complete control of the programmer, although a wide range of useful distributions is easily available through predefined functions. A crucial aspect of the library is that it does not allocate space for distributed arrays, but accepts programmer-specified memory. This has two major consequences. First, codes parallelized using Charon do not suffer from encapsulation; user data is always directly accessible. This provides high efficiency, and also retains the possibility of using message passing directly for highly irregular communications. Second, non-distributed arrays can be interpreted as (trivial) distributions in the Charon sense, which allows them to be mapped to truly distributed arrays, and vice versa. This is the mechanism that enables incremental parallelization. In this paper we provide a brief introduction of the library and then focus on the actual steps in the parallelization process, using some representative examples from, among others, the NAS Parallel Benchmarks. We show how a complicated two-dimensional pipeline-the prototypical non-data-parallel algorithm- can be constructed with ease. To demonstrate the flexibility of the library, we give examples of the stepwise, efficient parallel implementation of nonlocal boundary conditions common in aircraft simulations, as well as the construction of the sequence of grids required for multigrid.

VanderWijngaart, Rob F.↗
NASA NTRS1993

Parallel-Vector Algorithm For Rapid Structural Anlysis

New algorithm developed to overcome deficiency of skyline storage scheme by use of variable-band storage scheme. Exploits both parallel and vector capabilities of modern high-performance computers. Gives engineers and designers opportunity to include more design variables and constraints during optimization of structures. Enables use of more refined finite-element meshes to obtain improved understanding of complex behaviors of aerospace structures leading to better, safer designs. Not only attractive for current supercomputers but also for next generation of shared-memory supercomputers.

Agarwal, Tarun R.↗
DOE OSTI2026

Ground and excited state gradients with end-to-end differentiable semiempirical quantum chemistry

Accurate and efficient gradients of molecular energy with respect to nuclear degrees of freedom are essential for geometry optimization and molecular dynamics, including simulations that go beyond the Born–Oppenheimer regime. A common approach involves deriving analytical formulas for new electronic structure methods, which is often conceptually difficult and requires tedious coding. Here, we implement analytical, semi-numerical, and automatic differentiation (AD)-based gradient pathways for semiempirical Hamiltonian models in the PYSEQM software package, leveraging both graphics processing unit (GPU) and central processing unit (CPU) architectures. We further extend these capabilities to excited states calculated using the configuration interaction singles and time-dependent Hartree–Fock ansätze. We benchmark wall time, peak memory usage, and accuracy across three molecular families of varying chemical complexity, including systems of up to a thousand atoms. For ground-state simulations, analytical and AD gradients achieve near-identical GPU runtimes, while semi-numerical gradients are slower on GPU but remain competitive on CPU. For excited states, both analytical and custom AD approaches using implicit differentiation show similar performance and low memory requirements, whereas gradients with full AD are memory-limited. AD gradients match analytical ones in accuracy across all tested systems, aided by a quaternion-based diatomic frame rotation for two-center quantities that ensures smooth energy surfaces. Overall, automatic differentiation emerges as a practical alternative to analytical gradients in semiempirical quantum chemistry, offering high accuracy while allowing seamless integration in AI-driven workflows and popular packages, such as PyTorch and JAX. Our results provide actionable guidance for selecting optimal gradient strategies in large-scale ground- and excited-state molecular dynamics simulations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗