Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “concurrent”

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 163 records · Page 9

Locality Aware Concurrent Start for Stencil Applications

Stencil computations are at the heart of many physical simulations used in scientific codes. Thus, there exists a plethora of optimization efforts for this family of computations. Among these techniques, tiling techniques that allow concurrent start have proven to be very efficient in providing better performance for these critical kernels. Nevertheless, with many core designs being the norm, these optimization techniques might not be able to fully exploit locality (both spatial and temporal) on multiple levels of the memory hierarchy without compromising parallelism. It is no longer true that the machine can be seen as a homogeneous collection of nodes with caches, main memory and an interconnect network. New architectural designs exhibit complex grouping of nodes, cores, threads, caches and memory connected by an ever evolving network-on-chip design. These new designs may benefit greatly from carefully crafted schedules and groupings that encourage parallel actors (i.e. threads, cores or nodes) to be aware of the computational history of other actors in close proximity. In this paper, we provide an efficient tiling technique that allows hierarchical concurrent start for memory hierarchy aware tile groups. Each execution schedule and tile shape exploit the available parallelism, load balance and locality present in the given applications. We demonstrate our technique on the Intel Xeon Phi architecture with selected and representative stencil kernels. We show improvement ranging from 5.58% to 31.17% over existing state-of-the-art techniques.

Locality Aware execution, jagged tiling↗

Runtime Concurrency Control and Operation Scheduling for High Performance Neural Network Training

In this paper, we extend an existing runtime system (the TensorFlow runtime) to enable automatic concurrency control and scheduling of operations. We explore performance modeling to predict the performance of operations with various thread- level parallelism. Our performance model is highly accurate and lightweight. Leveraging the performance model, our runtime system employs a set of scheduling strategies that co-run opera- tions to improve hardware utilization and system throughput. Our runtime system demonstrates a significant performance benefit. Comparing with using the recommended configurations for concurrency control and operation scheduling in TensorFlow, our approach achieves 36% performance (execution time) im- provement on average (up to 49%) for four neural network models, and achieves high performance close to the optimal one manually obtained by the user.

Liu, Jiawen↗

Efficient Modeling of HIL Multi-Grid System for Scalability & Concurrency in CPS Security Testbed

Cyber-event-triggered power grid blackout compels utility operators to intensify cyber-aware and physics-constrained recovery and restoration process. Recently, coordinated cyber attacks on the Ukrainian grid witnessed such a cyber-event-triggered power system blackout. Various cyber-physical system (CPS) testbeds have attempted with multitude designs to analyze such interdependent events and evaluate remedy measures. However, resource constraints and modular integration designs have been significant barriers while modeling large-scale grid models (scalability) and multi-grid isolated models (concurrency) under a single real-time execution environment for the hardware-in-the-loop (HIL) CPS security testbeds. This paper proposes a meticulous design and effective modeling for simulating large-scale grid models and multi-grid isolated models in a HIL realtime digital simulator environment integrated with industry-grade hardware and software systems. We have used our existing HIL CPS security testbed to demonstrate scalability by the realtime performance of a Texas-2000 bus US synthetic grid model and concurrency by the real-time performance of simultaneous ten IEEE-39 bus grid models and an IEEE-118 bus grid model. The experiments demonstrated significant results by 100% realtime performance with zero overruns, low latency while receiving and executing control signals from SEL Relays via IEC-61850 protocol and low latency while computing and transmitting grid data streams including stability measures via IEEE C37.118 synchrophasor data protocol to SEL Phasor Data Concentrators.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Effect of MLC leaf width on treatment adaptation and accuracy for concurrent irradiation of prostate and pelvic lymph nodes

Purpose: The aim of the study was to evaluate the impact of multileaf collimator (MLC) leaf width on treatment adaptation and delivery accuracy for concurrent treatment of the prostate and pelvic lymph nodes with intensity modulated radiation therapy (IMRT). Methods: Seventy-five kilovoltage cone beam CTs (KV-CBCT) from six patients were included for this retrospective study. For each patient, three different IMRT plans were created based on a planning CT using three different MLC leaf widths of 2.5, 5, and 10 mm, respectively. For each CBCT, the prostate displacement was determined by a dual image registration. Adaptive plans were created by shifting selected MLC leaf pairs to compensate for daily prostate movements. To evaluate the impact of MLC leaf width on the adaptive plan for each daily CBCT, three MLC shifted plans were created using three different leaf widths of MLCs (a total of 225 adaptive treatment plans). Selective dosimetric endpoints for the tumor volumes and organs at risk (OARs) were evaluated for these adaptive plans. Using the planning CT from a selected patient, MLC shifted plans for three hypothetical longitudinal shifts of 2, 4, and 8 mm were delivered on the three linear accelerators to test the deliverability of the shifted plans and to compare the dose accuracy of the shifted plans with the original IMRT plans. Results: Adaptive plans from 2.5 and 5 mm MLCs had inadequate dose coverage to the prostate (D99 < 97%, or D{sub mean} < 99% of the planned dose) in 6%-8% of the fractions, while adaptive plans from 10 mm MLC led to inadequate dose coverage to the prostate in 25.3% of the fractions. The average V{sub 56Gy} of the prostate over the six patients was improved by 6.4% (1.6%-32.7%) and 5.8% (1.5%-35.7%) with adaptive plans from 2.5 and 5 mm MLCs, respectively, when compared with adaptive plans from 10 mm MLC. Pelvic lymph nodes were well covered for all MLC adaptive plans, as small differences were observed for D99, D{sub mean}, and V{sub 50.4Gy}. Similar OAR sparing could be achieved for the bladder and rectum with all three MLCs for treatment adaptation. The MLC shifted plans can be accurately delivered on all three linear accelerators with accuracy similar to their original IMRT plans, where gamma (3%/3 mm) passing rates were 99.6%, 93.0%, and 92.1% for 2.5, 5, and 10 mm MLCs, respectively. The percentages of pixels with dose differences between the measurement and calculation being less than 3% of the maximum dose were 85.9%, 82.5%, and 70.5% for the original IMRT plans from the three MLCs, respectively. Conclusions: Dosimetric advantages associated with smaller MLC leaves were observed in terms of the coverage to the prostate, when the treatment was adapted to account for daily prostate movement for concurrent irradiation of the prostate and pelvic lymph nodes. The benefit of switching the MLC from 10 to 5 mm was significant (p Much-Less-Than 0.01); however, switching the MLC from 5 to 2.5 mm would not gain significant (p= 0.15) improvement. IMRT plans with smaller MLC leaf widths achieved more accurate dose delivery.

61 RADIATION PROTECTION AND DOSIMETRY↗

WE-AB-202-02: Incorporating Regional Ventilation Function in Predicting Radiation Fibrosis After Concurrent Chemoradiotherapy for Lung Cancer

Purpose: To investigate the incorporation of pre-therapy regional ventilation function in predicting radiation fibrosis (RF) in stage III non-small-cell lung cancer (NSCLC) patients treated with concurrent thoracic chemoradiotherapy. Methods: 37 stage III NSCLC patients were retrospectively studied. Patients received one cycle of cisplatin-gemcitabine, followed by two to three cycles of cisplatin-etoposide concurrently with involved-field thoracic radiotherapy between 46 and 66 Gy (2 Gy per fraction). Pre-therapy regional ventilation images of the lung were derived from 4DCT via a density-change-based image registration algorithm with mass correction. RF was evaluated at 6-months post-treatment using radiographic scoring based on airway dilation and volume loss. Three types of ipsilateral lung metrics were studied: (1) conventional dose-volume metrics (V20, V30, V40, and mean-lung-dose (MLD)), (2) dose-function metrics (fV20, fV30, fV40, and functional mean-lung-dose (fMLD) generated by combining regional ventilation and dose), and (3) dose-subvolume metrics (sV20, sV30, sV40, and subvolume mean-lung-dose (sMLD) defined as the dose-volume metrics computed on the sub-volume of the lung with at least 60% of the quantified maximum ventilation status). Receiver operating characteristic (ROC) curve analysis and logistic regression analysis were used to evaluate the predictability of these metrics for RF. Results: In predicting airway dilation, the area under the ROC curve (AUC) values for (V20, MLD), (fV20, fMLD), and (sV20, and sMLD) were (0.76, 0.70), (0.80, 0.74) and (0.82, 0.80), respectively. The logistic regression p-values were (0.09, 0.18), (0.02, 0.05) and (0.004, 0.006), respectively. With regard to volume loss, the corresponding AUC values for these metrics were (0.66, 0.57), (0.67, 0.61) and (0.71, 0.69), and p-values were (0.95, 0.90), (0.43, 0.64) and (0.08, 0.12), respectively. Conclusion: The inclusion of regional ventilation function improved predictability of radiation fibrosis. Dose-subvolume metrics provided a promising method for incorporating functional information into the conventional dose-volume parameters for outcome assessment.

60 APPLIED LIFE SCIENCES↗

Magnetic properties and concurrence for fluid {sup 3}He on kagome lattice

We present the results of magnetic properties and entanglement for kagome lattice using Heisenberg model with two- and three-site exchange interactions in strong magnetic field. Kagome lattice correspond to the third layer of fluid {sup 3}He absorbed on the surface of graphite. The magnetic properties and concurrence as a measure of pairwise thermal entanglement are studied by means of variational mean-field like treatment based on Gibbs-Bogoliubov inequality. The system exhibits different magnetic behaviors depending on the values of the exchange parameters (J{sub 2}, J{sub 3}). We have obtained the magnetization plateaus at low temperatures. The central theme of the paper is comparing the entanglement and magnetic behavior for kagome lattice. We have found that in the antiferromagnetic region behavior of the concurrence coincides with the magnetic susceptibility one.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Concurrent Selective Lymph Node Radiotherapy and S-1 Plus Cisplatin for Esophageal Squamous Cell Carcinoma: A Phase II Study

Background: The efficacy, toxicity, and patterns of failure of esophageal squamous cell carcinoma (ESCC) treated with selective lymph node (SLN) conventional fraction radiotherapy (CFRT) and S-1 plus cisplatin (CDDP) were evaluated. Patients and Methods: 67 Patients with clinical stage II–IVa ESCC were enrolled. The total dose of SLN CFRT was 60 Gy in 30 fractions over 6 weeks. The first course of radiation covered the primary and metastatic regional tumors and high-risk lymph nodal regions, given at 2 Gy/fraction for a dose of 40 Gy. In the second course, CFRT was delivered to the boost volume for an additional 20 Gy in 10 days, using 2 Gy/fraction. Two cycles of chemotherapy were given at the beginning of radiotherapy. CDDP at 25 mg/m{sup 2}/day was given on days 1–3 and days 22–24, and S-1 at 80 mg/m{sup 2}/day on days 1–14 and days 22–35. Patients achieving objective response after concurrent chemoradiotherapy underwent two additional cycles of chemotherapy. Results: The objective response rate (ORR) was 82.5%. Grade 3 or 4 toxicities included leukopenia (23.8%), neutropenia (14.3%), thrombocytopenia (14.3%), hemoglobin (4.8%), gastrointestinal (12.7%), skin (1.6%), and esophagus fistula (1.6%). One patient died of severe pneumonia, and two died of late toxicity because of esophagus fistula. With median follow-up of 32 months, the overall survival (OS) and progression-free survival (PFS) at 1 year and 2 years were 81.0% and 73.0%, and 63.5% and 49.2%, respectively. Conclusions: SLN RT concurrent with S-1 plus CDDP may represent a better strategy for treatment of ESCC patients.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

CONCURRENT, CONDENSED STEIN VARIATIONAL GRADIENT DESCENT FOR UNCERTAINTY QUANTIFICATION OF NEURAL NETWORKS

In this work, we propose a Stein variational gradient descent (SVGD) method to concurrently sparsify, train, and provide uncertainty quantification (UQ) of a complexly parameterized model, such as a neural network (NN). It employs a graph reconciliation and condensation process to reduce complexity and increase similarity in the Stein ensemble of parameterizations. Therefore, the proposed concurrent, condensed SVGD (ccSVGD) method can provide UQ on parameters, not just outputs. Furthermore, the parameter reduction speeds up the convergence of the Stein gradient descent as it reduces the combinatorial complexity by aligning and differentiating the sensitivity to parameters. These properties are demonstrated with an illustrative example and an application to a mechanical response representation problem in solid mechanics.

42 ENGINEERING↗

Concurrent evaluation of autophagy induction and Burkholderia infection at the single cell level

Autophagy is a natural, regulated cellular process that "cleans up" cellular debris by degrading and recycling dysfunctional proteins. There is a high potential impact of exploiting the benefits of autophagy to complement existing treatments, but little has been done to date on bacterial pathogens of defense concern such as Burkholderia pseudomallei, a highly virulent Select Agent pathogen that is intrinsically resistant to most classes of antibiotics. Assessment of autophagy in the context of infection typically requires use of multiple technologies in combination (e.g., Western analysis paired with microscopy or flow cytometry) as applied to heterogeneous populations of cells. To address this, we have developed a dual target reporter cell line (RAW264.7 LC3-BFP:mPlum, GFP-RelA) that enables concurrent visualization of infection and autophagy induction. We assessed the effect of clinically approved small molecule inducers of autophagy on infection by Burkholderia thailendensis, a closely related but less virulent surrogate for B. pseudomallei. The reporter cells were first infected with a B. thailendensis strain that constitutively expresses GFP, then treated with one of four known autophagy inducers (rapamycin, niclosamide, bromhexine HC1, or flubendazole) for 4 hours. Confocal fluorescence imaging was used to quantify autophagy stimulation at the single cell level. Autophagy maturation was observed as a decrease in BFP LC3 puncta with a concurrent increase in mPlum LC3 puncta. B. thailendensis infection was assessed by monitoring translocation of GFP-RelA (an NFkB subunit) into the nucleus and through quantitating the intracellular bacterial presence in single cells. Preliminary results indicate that bromhexine HC1 and niclosamide may hinder B. thailendensis' ability to replicate intracellularly and reduce overall bacterial survival.

59 BASIC BIOLOGICAL SCIENCES↗

Automatically parallelizing batch inference on deep neural networks using Fiats and Fortran 2023 `do concurrent`

This paper introduces novel programming strategies that leverage features of the Fortran 2023 standard of the International Standards Organization (ISO) to automatically parallelize computations on deep neural networks. The paper focuses on the interplay of object-oriented, parallel, and functional programming paradigms in the Fiats deep learning library. We demonstrate how several infrequently used language features play a role in enabling efficient, parallel execution. Specifically, the ability to explicitly declare that a procedure is pure facilitates inference in the context of the language’s loop-parallelism construct `do concurrent`. Also, explicitly prohibiting the overriding of a parent type’s type-bound procedures eliminates the need for dynamic dispatch in performance-critical code. Finally, this paper uses batch inference calculations on a neural network surrogate for atmospheric aerosol dynamics to demonstrate that LLVM Flang compiler’s automatic parallelization of `do concurrent` achieves roughly the same performance and scalability as achieved by OpenMP compiler directives. We also demonstrate that double-precision inference costs 37–72% longer runtime than default-real precision with most values in the range 57-60%.

Rouson, Damian↗

Expectations of Future Natural Hazards in Human Adaptation to Concurrent Extreme Events in the Colorado River Basin

Human adaptation to climate change is the outcome of long-term decisions continuously made and revised by local communities. Adaptation choices can be represented by economic investment models in which the often large upfront cost of adaptation is offset by the future benefits of avoiding losses due to future natural hazards. In this context, we investigate the role that expectations of future natural hazards have on adaptation in the Colorado River basin of the USA. We apply an innovative approach that quantifies the impacts of changes in concurrent climate extremes, with a focus on flooding events. By including the expectation of future natural hazards in adaptation models, we examine how public policies can focus on this component to support local community adaptation efforts. Findings indicate that considering the concurrent distribution of several variables makes quantification and prediction of extremes easier, more realistic, and consequently improves our capability to model human systems adaptation. Hazard expectation is a leading force in adaptation. Even without assuming increases in exposure, the Colorado River basin is expected to face harsh increases in damage from flooding events unless local communities are able to incorporate climate change and expected increases in extremes in their adaptation planning and decision making.

54 ENVIRONMENTAL SCIENCES↗

Reducing Concurrency Bottlenecks in Parallel I/O Workloads

To enable high performance parallel checkpointing we introduced the Parallel Log Structured File System (PLFS). PLFS is middleware interposed on the file system stack to transform concurrent writing of one application file into many non-concurrently written component files. The promising effectiveness of PLFS makes it important to examine its performance for workloads other than checkpoint capture, notably the different ways that state snapshots may be later read, to make the case for using PLFS in the Exascale I/O stack. Reading a PLFS file involved reading each of its component files. In this paper we identify performance limitations on broader workloads in an early version of PLFS, specifically the need to build and distribute an index for the overall file, and the pressure on the underlying parallel file system's metadata server, and show how PLFS's decomposed components architecture can be exploited to alleviate bottlenecks in the underlying parallel file system.

COMPUTER ARCHITECTURE↗

Fast concurrent array-based stacks, queues and deques using fetch-and-increment-bounded, fetch-and-decrement-bounded and store-on-twin synchronization primitives

Implementation primitives for concurrent array-based stacks, queues, double-ended queues (deques) and wrapped deques are provided. In one aspect, each element of the stack, queue, deque or wrapped deque data structure has its own ticket lock, allowing multiple threads to concurrently use multiple elements of the data structure and thus achieving high performance. In another aspect, new synchronization primitives FetchAndIncrementBounded (Counter, Bound) and FetchAndDecrementBounded (Counter, Bound) are implemented. These primitives can be implemented in hardware and thus promise a very fast throughput for queues, stacks and double-ended queues.

97 MATHEMATICS AND COMPUTING↗

Vectorization of Random Number Generation and Reproducibility of Concurrent Particle Transport Simulation

Efficient random number generation with high quality statistical properties and exact reproducibility of Monte Carlo simulations are important requirements in many areas of computational science. VecRNG is a package providing pseudo-random number generation (pRNG) in the context of a new library VecMath. This library bundles up several general-purpose mathematical utilities, data structures, and algorithms having both SIMD and SIMT (GPUs) support based on VecCore.~Several state-of-the-art RNG algorithms are implemented as kernels supporting parallel generation of random numbers in scalar, vector, and Cuda workflows. In this report, we will present design considerations, implementation details, and computing performance of parallel pRNG engines on both CPU and GPU. Reproducibility of propagating multiple particles in parallel for HEP event simulation is demonstrated, using GeantV-based examples, for both sequential and fine-grain track-level concurrent simulation wor kflows. Strategies for efficient uses of vectorized pRNG and non-overlapping streams of random number sequences in concurrent computing environments is discussed as well.

Jun, S. Y.↗

Concurrent Conditions Access across Validity Intervals in CMSSW

The CMS software system, known as CMSSW, has a generalized conditions, calibration, and geometry data products system called the EventSetup. The EventSetup caches results of reading or calculating data products based on the 'interval of validity', IOV, which is based on the time period for which that data product is appropriate. With the original single threaded CMSSW framework, updating only on an IOV boundary meant we only required memory for a single data product of a given type at any time during the program execution. In 2016 CMS transitioned to using a multithreaded framework as a way to save on memory during processing. This was accomplished by amortizing the memory cost of EventSetup data products across multiple concurrent events. To initially accomplish that goal required synchronizing event processing across IOV boundaries, thereby decreasing the scalability of the system. In this presentation we will explain how we used 'limited concurrent task queues' to allow co ncurrent IOVs while still being able to limit the memory utilized. In addition, we will present performance measurements for both threading scalability of event throughtput and memory utilization.

Dagenhart, W.David↗

Radiotherapy With or Without Concurrent Chemotherapy for Lymph Node Recurrence After Radical Surgery of Thoracic Esophageal Squamous Cell Carcinoma

Purpose: To retrospectively compare the outcomes of patients with lymph node recurrence after radical surgery of esophageal cancer, when given radiotherapy with or without concurrent chemotherapy. Methods and Materials: Between January 1996 and December 2005, the data from 73 patients with lymph node recurrence after radical surgery of thoracic esophageal squamous cell carcinoma were retrospectively reviewed. The patients were separated into two groups: radiochemotherapy (RC, 31 patients) and radiotherapy alone (RA, 42 patients). Patients in the RC group received at least two cycles of 5-fluorouracil/cisplatin chemotherapy concurrently with radiotherapy. Results: The median duration of follow-up was 11 months (range, 2-48). The overall survival rate for all patients was 46.7% and 4.7% at 1 and 3 years, respectively. The median overall survival time was 9 months (95% confidence interval, 6.96-11.04) and 17 months (95% confidence interval, 13.61-20.39) for RA and RC groups, respectively. The survival rate at 1 and 3 years was 62.5% and 10.5% in the RC group and 33.8% and 0% in the RA group (p = .0049, log-rank test; hazard ratio for death, 0.52; 95% confidence interval, 0.30-0.92). Acute toxicities were more frequent in the RC group than in the RA group. No significant differences were found in the late toxicity profiles between the two groups. Conclusion: The results of the present retrospective analysis suggest that RC should be considered an effective and well-tolerated treatment of patients with thoracic esophageal squamous cell carcinoma and postoperative lymph node recurrence.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Multiscale Concurrent Atomistic-Continuum (CAC) modeling of multicomponent alloys

We report strengthening in complex multicomponent systems such as solid solution alloys is controlled primarily by the dynamic interactions between dislocation lines and heterogeneously distributed solute species. Modeling of extended defect length scales in such multicomponent systems becomes prohibitively expensive, motivating the development of reduced order approaches. This work explores the application of the Concurrent Atomistic-Continuum (CAC) method to model dislocation mobility in random alloys at extended length scales. By employing recently developed average-atom interatomic potentials, the average “bulk” material response in coarse-grained regions interacts with true random solute species in the atomistic-scale domain. We demonstrate that spurious stresses in domain resolution transition regions are eliminated entirely due to the CAC formulation. Simultaneously, the key details of local stress fluctuation due to randomness in the dislocation core region are captured, and fluctuating stress smoothly decays to the long-range dislocation stress field response. Dislocation mobility calculations, for line lengths over 400 nm, are computed as a function of alloy composition in the model FeNiCr system and compared to full molecular dynamics (MD). The results capture the composition-dependent trends, while reducing degrees of freedom by nearly 40%. This approach can be readily extended to any system described by an EAM potential and facilitates the study of large-scale defect dynamics in complex solute environments to support computational alloy design.

36 MATERIALS SCIENCE↗

Synchronous and Concurrent Multidomain Computing Method for Cloud Computing Platforms

We present a numerical method for synchronous and concurrent solution of transient elastodynamics problem where the computational domain is divided into subdomains that may reside on separate computational platforms. Here, this work employs the variational multiscale discontinuous Galerkin (VMDG) method to develop interdomain transmission conditions for transient problems. The fine-scale modeling concept leads to variationally consistent coupling terms at the common interfaces. The method admits a large class of time discretization schemes, and decoupling of the solution for each subdomain is achieved by selecting any explicit algorithm. Numerical tests with a manufactured solution problem show optimal convergence rates. The energy history in a free vibration problem is in agreement with that of the solution from a monolithic computational domain.

97 MATHEMATICS AND COMPUTING↗