Results for “Memory management” — Page 26

NASA NTRSDate not supplied

Document Classification Techniques for Aviation Letters of Agreement

Often when working with historic air traffic management (ATM) documents, it is helpful to classify them into specific categories. In this paper, we conduct a thorough review of natural language processing techniques to perform this classification task on Letters of Agreement (LOAs), technical aviation documents outlining rules for utilizing US airspace. We evaluate multiple techniques for representing the text in the documents as embeddings: unigram and bigram Term Frequency Inverse Document Frequency (TFIDF), Word2Vec, Doc2Vec, GloVe and RoBERTa. We investigate a wide range of classification models: K-Nearest Neighbors, Random Forest, Support Vector Machines (SVM), Logistic Regression, Naive Bayes, Feed-Forward Neural Network, Convolutional Neural Networks (CNNs) and Long-Short Term Memory (LSTM). By comparing the different methods, we found the best overall approach for our task was to use unigram TFIDF representations with SVM while also gaining insight into how the other methodologies performed on a small technical datasets.

ATM↗

NASA NTRSDate not supplied

Document Classification Techniques for Aviation Letters of Agreement

Often when working with historic air traffic management (ATM) documents, it is helpful to classify them into specific categories. In this paper, we conduct a thorough review of natural language processing techniques to perform this classification task on Letters of Agreement (LOAs), technical aviation documents outlining rules for utilizing US airspace. We evaluate multiple techniques for representing the text in the documents as embeddings: unigram and bigram Term Frequency Inverse Document Frequency (TFIDF), Word2Vec, Doc2Vec, GloVe and RoBERTa. We investigate a wide range of classification models: K-Nearest Neighbors, Random Forest, Support Vector Machines (SVM), Logistic Regression, Naive Bayes, Feed-Forward Neural Network, Convolutional Neural Networks (CNNs) and Long-Short Term Memory (LSTM). By comparing the different methods, we found the best overall approach for our task was to use unigram TFIDF representations with SVM while also gaining insight into how the other methodologies performed on a small technical datasets.

ATM↗

DOE OSTI2025

Computational Performance Bounds Prediction in Quantum Computing With Unstable Noise

Quantum computing has significantly advanced in recent years, boasting devices with hundreds of quantum bits (qubits), hinting at its potential quantum advantage over classical computing. Yet, noise in quantum devices poses significant barriers to realizing this supremacy. Understanding noise’s impact is crucial for reproducibility and application reuse; moreover, the next-generation quantum-centric supercomputing essentially requires efficient and accurate noise characterization to support system management (e.g., job scheduling), where ensuring correct functional performance (i.e., fidelity) of jobs on available quantum devices can even be higher-priority than traditional objectives. However, noise fluctuates over time, even on the same quantum device, which makes predicting the computational bounds for on-the-fly noise is vital. Noisy quantum simulation can offer insights but faces efficiency and scalability issues. Here, in this work, we propose a data-driven workflow, namely QuBound, to predict computational performance bounds. It decomposes historical performance traces to isolate noise sources and devises a novel encoder to embed circuit and noise information processed by a Long Short-Term Memory (LSTM) network. For evaluation, we compare QuBound with a state-of-the-art learning-based predictor, which only generates a single performance value instead of a bound. Experimental results show that the result of the existing approach falls outside of performance bounds, while all predictions from our QuBound with the assistance of performance decomposition better fit the bounds. Moreover, QuBound can efficiently produce practical bounds for various circuits with over 106 speedup over simulation; in addition, the range from QuBound is over 10× narrower than the state-of-the-art analytical approach.

Li, Jinyang [George Mason Univ., Fairfax, VA (Unit↗

NASA NTRS1990

Multi-processing on supercomputers for computational aerodynamics

The MIMD concept is applied, through multitasking, with relatively minor modifications to an existing code for a single processor. This approach maps the available memory to multiple processors, exploiting the C-FORTRAN-Unix interface. An existing single processor algorithm is mapped without the need for developing a new algorithm. The procedure of designing a code utilizing this approach is automated with the Unix stream editor. A Multiple Processor Multiple Grid (MPMG) code is developed as a demonstration of this approach. This code solves the three-dimensional, Reynolds-averaged, thin-layer and slender-layer Navier-Stokes equations with an implicit, approximately factored and diagonalized method. This solver is applied to a generic, oblique-wing aircraft problem on a four-processor computer using one process for data management and nonparallel computations and three processes for pseudotime advance on three different grid systems.

Yarrow, Maurice↗

NASA NTRS2000

Free-Flying Magnetometer Data System

The Free-Flying Magnetometer (FFM) is an autonomous "sensorcraft" developed at the Jet Propulsion Laboratory (JPL) for the Enstrophy sounding rocket mission. This mission was a collaborative project between the University of New Hampshire, Cornell University and JPL. The science goal of the mission was the study of current filamentation phenomena in the northern auroral region through multipoint measurements of magnetic field. The technical objective of the mission was the proof of concept of the JPL FFM design and the demonstration of an in-situ multipoint measurement technique employing many free-flying spacecraft. Four FFMs were successfully deployed from a sounding rocket launched from Poker Flats, Alaska on February 11, 1999. These hockey-puck-sized (80 mm diameter, 38 mm. height, 250 gram mass) free flyers each carry a miniature 3-axis flux-gate magnetometer that output +/- 2 V signals corresponding to a +/- 60,000 nT measurement range for each axis. The FFM uses a synchronized four-channel Sigma(Delta) Analog-to-Digital Converter (ADC) having a dynamic range of +/- 2.5V and converting at a rate of 279 samples/second/channel. Three channels are used to digitize the magnetometer signals to 17-bit (1.144 nT/bit) resolution. The fourth ADC channel is multiplexed for system monitoring of four temperature sensors and two battery voltages. The FFM also contains two sun sensors, a laser diode which emits a fan-shaped beam, a miniature S-band transmitter for direct communication to the ground station antennas, an ultra-stable Temperature Compensated Crystal Oscillator (TCXO) clock, an integrated data subsystem implemented in a Field-Programmable Gate Array (FPGA), a 4 Mbit Static Random Access Memory (SRAM) for data storage and Lithium Thionyl Chloride batteries for power. Communicating commands to the FFM prior to deployment is achieved with an infrared (IR) link. The FFM IR receiver responds to 9-bit pulse coded signals that are generated by an IR Light Emitting Diode (LED) in the payload for turning FFM power on or off and placing the FFM in a test mode or flight mode. The IR links are also used to synchronize (zero) the clocks onboard all the FFMs through a reset pulse originating from the payload GPS receiver that is issued when the FFMs are in flight mode. The FPGA based data subsystem manages continuous data collection from the four ADC channels and sun sensors, formatting and storing the data to SRAM, and controlling downlink transmission. The transmitter is powered only after a 2547 frame SRAM buffer has been filled (approx. 5 minutes of data). The data is Viterbi encoded and sent to the S-band transmitter via a First-In-First-Out (FIFO) buffer who's output is clocked at 100 bits/second. After the 26-second transmission, the transmitter is turned off to reduce noise coupling to the sensitive magnetometer. The data subsystem control consists of a master state machine that performs data flow management and is interfaced through a prioritized interrupt scheme to state machines that service the ADC, sun sensors and transmitter FIFO. Continuous data collection prevents the missing of data during transmission and provides implicit time tagging of the data acquired by the ADC because of synchronization with the TCXO clock.

Blaes, B.↗

DOE OSTI2023

Explainable machine learning model for multi-step forecasting of reservoir inflow with uncertainty quantification

We propose an explainable machine learning (ML) model with uncertainty quantification (UQ) to improve multi-step reservoir inflow forecasting. Traditional ML methods have challenges in forecasting inflows multiple days ahead, and lack explainability and UQ. To address these limitations, we introduce an encoder–decoder long short-term memory (ED-LSTM) network for multi-step forecasting, employ the SHapley Additive exPlanation (SHAP) technique for understanding the influence of hydrometeorological factors on inflow prediction, and develop a novel UQ method for prediction trustworthiness. We apply these methods to forecast 7-day inflow in snow-dominant and rain-driven reservoirs. The results demonstrate the effectiveness of the ED-LSTM model, with high forecasting accuracy for short lead times. Our UQ method provides reliable uncertainty estimates, covering 90% of data with a 90% confidence level. The SHAP analysis reveals the importance of historical inflow and precipitation as influential factors. These findings and methods may support reservoir operators in optimizing water resources management decisions.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS2002

An Efficient Scheme for Updating Sparse Cholesky Factors

Raghavan had earlier developed the software package DCSPACK which can be used for solving sparse linear systems where the coefficient matrix is symmetric and positive definite (this project was not funded by NASA but by agencies such as NSF). DSCPACK-S is the serial code and DSCPACK-P is a parallel implementation suitable for multiprocessors or networks-of-workstations with message passing using MCI. The main algorithm used is the Cholesky factorization of a sparse symmetric positive positive definite matrix A = LL(T). The code can also compute the factorization A = LDL(T). The complexity of the software arises from several factors relating to the sparsity of the matrix A. A sparse N x N matrix A has typically less that cN nonzeroes where c is a small constant. If the matrix were dense, it would have O(N2) nonzeroes. The most complicated part of such sparse Cholesky factorization relates to fill-in, i.e., zeroes in the original matrix that become nonzeroes in the factor L. An efficient implementation depends to a large extent on complex data structures and on techniques from graph theory to reduce, identify, and manage fill. DSCPACK is based on an efficient multifrontal implementation with fill-managing algorithms and implementation arising from earlier research by Raghavan and others. Sparse Cholesky factorization is typically a four step process: (1) ordering to compute a fill-reducing numbering, (2) symbolic factorization to determine the nonzero structure of L, (3) numeric factorization to compute L, and, (4) triangular solution to solve L(T)x = y and Ly = b. The first two steps are symbolic and are performed using the graph of the matrix. The numeric factorization step is of dominant cost and there are several schemes for improving performance by exploiting the nested and dense structure of groups of columns in the factor. The latter are aimed at better utilization of the cache-memory hierarchy on modem processors to prevent cache-misses and provide execution rates (operations/second) that are close to the peak rates for dense matrix computations. Currently, EPISCOPACY is being used in an application at NASA directed by J. Newman and M. James. We propose the implementation of efficient schemes for updating the LL(T) or LDL(T) factors computed in DSCPACK-S to meet the computational requirements of their project. A brief description is provided in the next section.

Raghavan, Padma↗

NASA NTRS2007

[Interview Questions]

The Goddard Mission Services Evolution Center, or GMSEC, was started in 2001 to create a new standard approach for managing GSFC missions. Standardized approaches in the past involved selecting and then integrating the most appropriate set of functional tools. Assumptions were made that "one size fits all" and that tool changes would not be necessary for many years. GMSEC took a very different approach and has proven to be very successful. The core of the GMSEC architecture consists of a publish/subscribe message bus, standardized message formats, and an Applications Programming Interface (API). The API supports multiple operating systems, programming languages and messaging middleware products. We use a GMSEC-developed free middleware for low-cost development. A high capacity, robust middleware is used for operations and a messaging system with a very small memory footprint is used for on-board flight software. Software components can use the standard message formats or develop adapters to convert from their native formats to the GMSEC formats. We do not want vendors to modify their core products. Over 50 software components are now available for use with the GMSEC architecture. Most available commercial telemetry and command systems, including the GMV hifly Satellite Control System, have been adapted to run in the GMSEC labs.

Smith, Dan↗

DOE OSTI2021

Athena: High-Performance Sparse Tensor Contraction Sequence on Heterogeneous Memory

Sparse tensor contraction sequence has been widely employed in many fields, such as chemistry and physics. However, how to efficiently implement the sequence faces multiple challenges, such as redundant computations and memory operations, massive memory consumption, and inefficient utilization of hardware. To address the above challenges, we introduce Athena, a high-performance framework for SpTC sequences. Athena introduces new data structures, leverages emerging Optane-based heterogeneous memory (HM) architecture, and stage parallelism. In particular, Athena introduces shared hash table-represented sparse accumulator to eliminate unnecessary input processing and data migration; Athena uses a novel data-semantic guided dynamic migration solution to make the best use of the Optane-based HM for high performance; Athena also co-runs execution phases with different characteristics to enable high hardware utilization. Evaluating with 12 datasets, we show that Athena brings 327-7362× speedup over the state-of-the-art SpTC algorithm. With the dynamic data placement guided by data semantics, Athena brings performance improvement on Optane-based HM over a state-of-the-art software-based data management solution, a hardware-based data management solution, and PMM-only by 1.58×, 1.82×, and 2.34× respectively.

Liu, Jiawen↗

NASA NTRS2011

NASA Tech Briefs, September 2011

Topics covered include: Fused Reality for Enhanced Flight Test Capabilities; Thermography to Inspect Insulation of Large Cryogenic Tanks; Crush Test Abuse Stand; Test Generator for MATLAB Simulations; Dynamic Monitoring of Cleanroom Fallout Using an Air Particle Counter; Enhancement to Non-Contacting Stress Measurement of Blade Vibration Frequency; Positively Verifying Mating of Previously Unverifiable Flight Connectors; Radiation-Tolerant Intelligent Memory Stack - RTIMS; Ultra-Low-Dropout Linear Regulator; Excitation of a Parallel Plate Waveguide by an Array of Rectangular Waveguides; FPGA for Power Control of MSL Avionics; UAVSAR Active Electronically Scanned Array; Lockout/Tagout (LOTO) Simulator; Silicon Carbide Mounts for Fabry-Perot Interferometers; Measuring the In-Process Figure, Final Prescription, and System Alignment of Large; Optics and Segmented Mirrors Using Lidar Metrology; Fiber-Reinforced Reactive Nano-Epoxy Composites; Polymerization Initiated at the Sidewalls of Carbon Nanotubes; Metal-Matrix/Hollow-Ceramic-Sphere Composites; Piezoelectrically Enhanced Photocathodes; Iridium-Doped Ruthenium Oxide Catalyst for Oxygen Evolution; Improved Mo-Re VPS Alloys for High-Temperature Uses; Data Service Provider Cost Estimation Tool; Hybrid Power Management-Based Vehicle Architecture; Force Limit System; Levitated Duct Fan (LDF) Aircraft Auxiliary Generator; Compact, Two-Sided Structural Cold Plate Configuration; AN Fitting Reconditioning Tool; Active Response Gravity Offload System; Method and Apparatus for Forming Nanodroplets; Rapid Detection of the Varicella Zoster Virus in Saliva; Improved Devices for Collecting Sweat for Chemical Analysis; Phase-Controlled Magnetic Mirror for Wavefront Correction; and Frame-Transfer Gating Raman Spectroscopy for Time-Resolved Multiscalar Combustion Diagnostics.

Source record↗

DOE OSTI2022

Cryogenic energy storage: Standalone design, rigorous optimization and techno-economic analysis

Energy storage allows flexible use and management of excess electricity and intermittently available renewable energy. Cryogenic energy storage (CES) is a promising storage alternative with a high technology readiness level and maturity, but the round-trip efficiency is often moderate and the Levelized Cost of Storage (LCOS) remains high. The complex flowsheets with intricate thermodynamics at cryogenic temperatures as well as the presence of multiple loops and refrigeration cycles pose considerable challenges for rigorous model-based design and optimization of CES systems. We present an optimization strategy that couples rigorous process simulation and Bayesian optimization with flowsheet decomposition and identification of hidden coupling constraints to optimally design standalone CES systems. Further refinement is done via a local search using the limited-memory Broyden–Fletcher–Goldfarb–Shanno algorithm. Here our results indicate that it is possible to achieve more than 52% round-trip efficiency and an LCOS of $153/MWh for a standalone 100 MW/400 MWh CES system limited to short-term storage with daily charging–discharging. However, a detailed techno-economic assessment reveals that the LCOS considering total capital investment may exceed $267/MWh when all direct and indirect costs of installation and operation are considered.

25 ENERGY STORAGE↗

NASA NTRS1990

Galileo spacecraft power management and distribution system

The Galileo PMAD (power management and distribution system) is described, and the design drivers that established the final as-built hardware are discussed. The spacecraft is powered by two general-purpose heat-source-radioisotope thermoelectric generators. Power bus regulation is provided by a shunt regulator. Galileo PMAD distributes a 570-W beginning of mission (BOM) power source to a user complement of some 137 load elements. Extensive use of pyrotechnics requires two pyro switching subassemblies. They initiate 148 squibs which operate the 47 pyro devices on the spacecraft. Detection and correction of faults in the Galileo PMAD is an autonomous feature dictated by requirements for long life and reliability in the absence of ground-based support. Volatile computer memories in the spacecraft command and data system and attitude control system require a continuous source of backup power during all anticipated power bus fault scenarios. Power for the Jupiter Probe is conditioned, isolated, and controlled by a Probe interface subassembly. Flight performance of the spacecraft and the PMAD has been successful to date, with no major anomalies.

Detwiler, R. C.↗

DOE OSTI2022

Determinants of residential mobility: an adaptive retrospective survey method

This study introduces a survey instrument to collect retrospective life-course events, focusing on residential relocation, and it utilizes the survey to evaluate determinants of residential mobility. Here, the survey consists of seven modules collecting information about household structure and household demographics, latest residential relocation, current and previous home, employment, education, vehicle ownership, and travel behavior. The time window of the life-course calendar in the survey is customized according to the latest residential relocation as an anchor point to assist with memory recollection and balance the required input from participants. The survey is used to collect data from a sample of 514 respondents in Sydney, Australia, and another sample of 404 respondents in Chicago, Illinois, in the US. The Cox proportional hazard model is used to analyze residential mobility. The results show primary school commencement is a salient determinant of residential relocation, but its impact is significantly higher in Chicago compared to Sydney.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

NASA NTRS1996

Report on Alternative Devices to Pyrotechnics on Spacecraft

Pyrotechnics accomplish many functions on today's spacecraft, possessing minimum volume/weight, providing instantaneous operation on demand, and requiring little input energy. However, functional shock, safety, and overall system cost issues, combined with emergence and availability of new technologies question their continued use on space missions. Upon request from the National Aeronautics and Space Administration's (NASA) Program Management Council (PMC), Langley Research Center (LaRC) conducted a survey to identify and evaluate state-of-the-art non-explosively actuated (NEA) alternatives to pyrotechnics, identify NEA devices planned for NASA use, and investigate potential interagency cooperative efforts. In this study, over 135 organizations were contacted, including NASA field centers, Department of Defense (DOD) and other government laboratories, universities, and American and European industrial sources resulting in further detailed discussions with over half, and 18 face-to-face briefings. Unlike their single use pyrotechnic predecessors, NEA mechanisms are typically reusable or refurbishable, allowing flight of actual tested units. NEAs surveyed include spool-based devices, thermal knife, Fast Acting Shockless Separation Nut (FASSN), paraffin actuators, and shape memory alloy (SMA) devices (e.g., Frangibolt). The electro-mechanical spool, paraffin actuator and thermal knife are mature, flight proven technologies, while SMA devices have a limited flight history. There is a relationship between shock, input energy requirements, and mechanism functioning rate. Some devices (e.g., Frangibolt and spool based mechanisms) produce significant levels of functional shock. Paraffin, thermal knife, and SMA devices can provide gentle, shock-free release but cannot perform critically timed, simultaneous functions. The FASSN flywheel-nut release device possesses significant potential for reducing functional shock while activating nearly instantaneously. Specific study recommendations include: (1) development of NEA standards, specifically in areas of material characterization, functioning rates, and test methods; (2) a systems level approach to assure successful NEA technology application; and (3) further investigations into user needs, along with industry/government system-level real spacecraft cost benefit trade studies to determine NEA application foci and performance requirements. Additional survey observations reveal an industry and government desire to establish partnerships to investigate remaining unknowns and formulate NEA standards, specifically those driven by SMAs. Finally, there is increased interest and need to investigate alternative devices for such functions as stage/shroud separation and high pressure valving. This paper summarizes results of the NASA-LaRC survey of pyrotechnic alternatives. State of-the-art devices with their associated weight and cost savings are presented. Additionally, a comparison of functional shock characteristics of several devices are shown, and potentially related technology developments are highlighted.

Lucy, M. H.↗

DOE OSTI2023

UPC++ v1.0 Programmer’s Guide (Rev. 2023.9.0)

UPC++ is a C++ library that supports Partitioned Global Address Space (PGAS) programming. It is designed for writing efficient, scalable parallel programs on distributed-memory parallel computers. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). The UPC++ control model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. The PGAS memory model additionally provides one-sided RMA communication to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. UPC++ was designed to support exascale high-performance computing, and the library interfaces and implementation are focused on maximizing scalability. In UPC++, all communication operations are syntactically explicit, which encourages programmers to consider the costs associated with communication and data movement. Moreover, all communication operations are asynchronous by default, encouraging programmers to seek opportunities for overlapping communication latencies with other useful work. UPC++ provides expressive and composable abstractions designed for efficiently managing aggressive use of asynchrony in programs. Together, these design principles are intended to enable programmers to write applications using UPC++ that perform well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

UPC++ v1.0 Programmer’s Guide (Revision 2022.3.0)

UPC++ is a C++ library that supports Partitioned Global Address Space (PGAS) programming. It is designed for writing efficient, scalable parallel programs on distributed-memory parallel computers. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). The UPC++ control model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. The PGAS memory model additionally provides one-sided RMA communication to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. UPC++ was designed to support exascale high-performance computing, and the library interfaces and implementation are focused on maximizing scalability. In UPC++, all communication operations are syntactically explicit, which encourages programmers to consider the costs associated with communication and data movement. Moreover, all communication operations are asynchronous by default, encouraging programmers to seek opportunities for overlapping communication latencies with other useful work. UPC++ provides expressive and composable abstractions designed for efficiently managing aggressive use of asynchrony in programs. Together, these design principles are intended to enable programmers to write applications using UPC++ that perform well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

UPC++ v1.0 Programmer’s Guide, Revision 2023.3.0

UPC++ is a C++ library that supports Partitioned Global Address Space (PGAS) programming. It is designed for writing efficient, scalable parallel programs on distributed-memory parallel computers. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). The UPC++ control model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. The PGAS memory model additionally provides one-sided RMA communication to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. UPC++ was designed to support exascale high-performance computing, and the library interfaces and implementation are focused on maximizing scalability. In UPC++, all communication operations are syntactically explicit, which encourages programmers to consider the costs associated with communication and data movement. Moreover, all communication operations are asynchronous by default, encouraging programmers to seek opportunities for overlapping communication latencies with other useful work. UPC++ provides expressive and composable abstractions designed for efficiently managing aggressive use of asynchrony in programs. Together, these design principles are intended to enable programmers to write applications using UPC++ that perform well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

UPC++ v1.0 Programmer’s Guide, Revision 2022.9.0

UPC++ is a C++ library that supports Partitioned Global Address Space (PGAS) programming. It is designed for writing efficient, scalable parallel programs on distributed-memory parallel computers. The key communication facilities in UPC++ are one-sided Remote Memory Access (RMA) and Remote Procedure Call (RPC). The UPC++ control model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. The PGAS memory model additionally provides one-sided RMA communication to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. UPC++ was designed to support exascale high-performance computing, and the library interfaces and implementation are focused on maximizing scalability. In UPC++, all communication operations are syntactically explicit, which encourages programmers to consider the costs associated with communication and data movement. Moreover, all communication operations are asynchronous by default, encouraging programmers to seek opportunities for overlapping communication latencies with other useful work. UPC++ provides expressive and composable abstractions designed for efficiently managing aggressive use of asynchrony in programs. Together, these design principles are intended to enable programmers to write applications using UPC++ that perform well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗