Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed memory algorithm”

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 559 records · Page 31

SMC 2021 Data Challenge: Analyzing Resource Utilization and User Behavior on Titan Supercomputer

Resource utilization statistics of submitted jobs on a supercomputer can help us understand how users from various scientific domains use HPC platforms and better design a job scheduler. We explore to generate insight regarding workload distribution and usage pattern domains from job scheduler trace, GPU failure information, and project-specific information collected from Titan supercomputer. Furthermore, we want to know how the scheduler performance varies over time and how the users' scheduling behavior changes following a system failure. These observations have the potential to provide valuable insight, which is helpful to prepare for system failures. These practices will help us develop and apply novel machine learning algorithms in understanding system behavior, requirement, and better scheduling of HPC systems. There are two datasets, RUR and GPU: RUR dataset is the job scheduler traces collected from the Titan supercomputer from 01/01/2015 to 07/31/2019 (2015.csv - 2019.csv). These were collected using resource Utilization Report (RUR), a Cray-developed resource-usage data collection and reporting system. It contains the usage information of its critical resources (CPU, Memory, GPU, and I/O) of each running job on Titan during that period (https://ieeexplore.ieee.org/abstract/document/8891001). It includes ProjectAreas as additional information, every job is associated with a project ID. TheProjectAreas.csv dataset provides a mapping of the project ID to its domain science. GPU dataset has information regarding GPU failure on Titan. There have been some hardware-related issues in the GPUs in Titan that caused some GPUs to fail, sometimes irrecoverably during some job runs. This dataset provides information regarding these failures during the execution of the submitted jobs. GPUs on Titan are uniquely identified by a serial number (SN), and they are installed in a location. A GPU can be installed in a location, then removed from that location following a failure, and then re-installed in a different location after fixing the problem. If the failure can't be recovered, the GPU might be removed entirely from Titan. There are two prominent types of failures that resulted in the removal of GPUs from Titan: Double Bit Error (DBE) and Out of the Bus (OTB). The dataset (gc_full.csv) has seven attributes, we provided a short description of these attributes in the ReadMe file. To learn more about this dataset, please refer to the git repository https://github.com/olcf/TitanGPULife and the related publication (https://ieeexplore.ieee.org/abstract/document/9355319).

42 ENGINEERING↗

DIMPLES: Distributed Influence Maximization for Pandemic pLanning on Exascale Systems

We study exascale parallel algorithms for the selection of intervention or monitoring strategies in massive realistic socio-technical networks through scalable Influence Maximization (InfMax) algorithms. We employ novel techniques to enable efficient scaling on up to 8k nodes of OLCF Frontier, with 65k AMD GPUs and 458k AMD CPU cores. Current state-of-the-art InfMax tools are limited to networks with only a few million actors (vertices) and a few hundred million interactions (edges). By overcoming these limitations, we show that our approach is capable of processing a realistic social contact network of the United States with 285 million nodes and about 8 billion edges. This two orders-of-magnitude improvement over the previous state-of-the-art is obtained by leveraging algorithmic advancements for the InfMax problem and designing several problem-specific approaches to overlap communication with computation, improve GPU efficiency, and lower the application’s memory requirements. We evaluate strong scaling for computing 10k most influential seeds using up to 8k nodes of an exascale system, and weak scaling from 128 to 8k system nodes for seed sets ranging from 625 to 40k seeds. We achieve the fastest-known runtime of 25 minutes while performing 48 million diffusion simulations totaling 2.31 petabytes to identify 40k influential seeds using 8k nodes, and take 5.75 minutes to identify 10k seeds while using 4k nodes.

Minutoli, Marco [Pacific Northwest National Labora↗

Security Enhancement of Network Constraint Grid-Edge Energy Management System

Network constrained grid edge energy management system (EMS) provides economic solution for active and reactive power dispatch of distributed energy resources (DERs) at the grid edge level. Grid edge EMS ensures secure interconnection of a circuit segment to the distribution system by maintaining grid code requirements (e.g. IEEE 1547–2018). Grid edge EMS is dependent on communication to receive load measurement, which brings a risk of unobservable false data injection attacks (FDIAs). To mitigate the risk, this paper proposes a framework to enhance resilient operation of grid edge EMS by detecting the unobservable FDIAs on loads and replacing them with forecasted values. In this work, a two-step detection algorithm is proposed. In first step, conventional residual based algorithm is deployed. Autoencoder (AE) based data driven mechanism is included in second step to detect the presence of unobservable FDIAs. After ensuring the presence of FDIA, its specific location is detected by checking the maximum residue values till the predefined threshold value is reached. Detected false data injected loads are then replaced with forecasted load values following long-short term memory (LSTM) based forecast to ensure resilient performance of grid edge EMS in the presence of attacks. This proposed security enhancement framework for grid edge EMS is evaluated in IEEE 13 bus system with three integrated DERs. Numerical simulation shows the validation of the proposed framework by reducing voltage violation in real operation of grid edge EMS.

cyber attack detection↗

toolkit computational mesh conceptual model.

The Sierra Toolkit computational mesh is a software library intended to support massively parallel multi-physics computations on dynamically changing unstructured meshes. This domain of intended use is inherently complex due to distributed memory parallelism, parallel scalability, heterogeneity of physics, heterogeneous discretization of an unstructured mesh, and runtime adaptation of the mesh. Management of this inherent complexity begins with a conceptual analysis and modeling of this domain of intended use; i.e., development of a domain model. The Sierra Toolkit computational mesh software library is designed and implemented based upon this domain model. Software developers using, maintaining, or extending the Sierra Toolkit computational mesh library must be familiar with the concepts/domain model presented in this report.

97 MATHEMATICS AND COMPUTING↗

CRIC and Rucio integration for CMS Data Management

The Large Hadron Collider (LHC) at CERN smashes protons together at close to the speed of light with seven times the energy of the most powerful accelerators built up to now. Some of the collision energy is turned into mass, creating new particles which are observed in the Compact Muon Solenoid (CMS) particle detector. CMS data is analyzed by scientists around the world to build up a picture of what happened at the heart of the collision. This will help us answer questions such as: ”what is the Universe really made of and what forces act within it?” and ”what gives everything substance?”. All the data produced by the experiment are collected and stored in a distributed system, since it must provide an high throughput and performing access to these data to scientists and physicists spread all over the world. In order to answer the previous questions and many others, CMS users must first obtain a copy of a portion of these data, and then to analyze it, therefore the first challenge is to decide in which storage site to save this duplicate so they can access it as quickly as possible. The second issue is each storage site shares a limited memory space with many CMS users, and each one needs to be assigned to a well defined personal amount of storage to pin data he needs. The main goal of my Summer Internship is to make scalable and automatic this user-site mapping which is based on a preexisting policy, proving a structured and organized algorithm to accomplish this task. Hence we are going to define, and solve, the main challenges encountered, and the technical solution used to generalize the algorithm to handle different policies for different countries.

43 PARTICLE ACCELERATORS↗

High Order Anchoring and Reinitialization of Level Set Function for Simulating Interface Motion

A second order interface anchoring method has been developed and used with fast sweeping algorithm for reinitialization of a level set function. The algebraic anchoring formulation ensures that the location of the actual interface is preserved, leading to better mass conservation property. It also provides high order accurate algebraic constraint for solving the Eikonal equation on a finite difference grid. Geometric properties of the interface such as surface normal and curvature are subsequently computed from the reinitialized distance function. Various analytical functions for modeling distortion in level set field are considered and accuracy of reinitialization is evaluated using first and second order anchoring schemes. It is also shown that accurate computation of interface curvature requires a high order anchor in addition to a high order fast sweeping method. Mass conservation property of reinitialization is also analyzed by considering test problems from literature including the classic Rider–Kothe single vortex problem. The formulation is suitable for efficient parallelization for both distributed memory and on-node shared memory parallel systems. In conclusion, scalability and performance of the reinitialization scheme on multiple architectures are demonstrated.

97 MATHEMATICS AND COMPUTING↗

Using parallel banded linear system solvers in generalized eigenvalue problems

Subspace iteration is a reliable and cost effective method for solving positive definite banded symmetric generalized eigenproblems, especially in the case of large scale problems. This paper discusses an algorithm that makes use of two parallel banded solvers in subspace iteration. A shift is introduced to decompose the banded linear systems into relatively independent subsystems and to accelerate the iterations. With this shift, an eigenproblem is mapped efficiently into the memories of a multiprocessor and a high speedup is obtained for parallel implementations. An optimal shift is a shift that balances total computation and communication costs. Under certain conditions, we show how to estimate an optimal shift analytically using the decay rate for the inverse of a banded matrix, and how to improve this estimate. Computational results on iPSC/2 and iPSC/860 multiprocessors are presented.

DISTRIBUTED MEMORY MULTIPROCES↗

Power inverter implementing phase skipping control

A power inverter includes a DC/AC inverter having first, second and third phase circuitry coupled to receive power from a power source. A controller is coupled to a driver for each of the first, second and third phase circuitry (control input drivers). The controller includes an associated memory storing a phase skipping control algorithm, wherein the controller is coupled to receive updating information including a power level generated by the power source. The drivers are coupled to control inputs of the first, second and third phase circuitry, where the drivers are configured for receiving phase skipping control signals from the controller and outputting mode selection signals configured to dynamically select an operating mode for the DC/AC inverter from a Normal Control operation and a Phase Skipping Control operation which have different power injection patterns through the first, second and third phase circuitry depending upon the power level.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Framework for discrete-time quantum walks and a symmetric walk on a binary tree

We formulate a framework for discrete-time quantum walks, motivated by classical random walks with memory. We present a specific representation of the classical walk with memory 2, on which this is based. The framework has no need for coin spaces, it imposes no constraints on the evolution operator other than unitarity, and is unifying of other approaches. As an example we construct a symmetric discrete-time quantum walk on the semi-infinite binary tree. The generating function of the amplitude at the root is computed in closed form, as a function of time and the initial level n in the tree, and we find the asymptotic and a full numerical solution for the amplitude. It exhibits a sharp interference peak and a power-law tail, as opposed to the exponentially decaying tail of a broadly peaked distribution of the classical symmetric random walk on a binary tree. The probability peak is orders of magnitude larger than it is for the classical walk (already at small n). The quantum walk shows a polynomial algorithmic speedup in n over the classical walk, which we conjecture to be of the order 2/3, based on strong trends in data.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

EmSense: A High-Resolution Emulated Sensor for Experiments with the Smart Grid and Distributed Ledger Technology

This work involves the development of a device - EmSense (“Emulated Sensor”) - that emulates a high-resolution sensor for a power grid. The device collects raw current and voltage sensor data which derive from ORNL's signature library. This library is a dataset that ORNL curates from many different sources that include power systems from various utilities. The EmSense packages the data from the library in the form of IEC 61850 Sampled Value (SV) packets and then broadcasts these SV packets on the network. In another mode, EmSense can generate artificial sinusoidal data that appears as waveforms for voltage and current signals. EmSense has an internal algorithm for determining the period of a signal based on the data so that the period can be specified as a variable in the IEC 61850 packets. The purpose of EmSense is to allow for experimentation with the Dark Net Infrastructure where a variety of power line sensors must be represented along with their typical communication traffic. The EmSense device was developed in coordination with the software for receiving and processing the packets in the Distributed Ledger Technology (DLT) framework of the DarkNet Project. This receiving software must have a methodology for dealing with information of high velocity, variety, and volume. Experimenting with EmSense facilitates the development of such software. The results showed that the DLT framework and the trust-anchoring approach managed to process a large flow of traffic even with up to six instances of EmSense device broadcasting data. This was achieved without overfilling packet queues in the memory of the actual hardware of the DLT devices or causing the Central Processing Unit (CPU) of the hardware to be overwhelmed. The DLTs were also able to store the data in a compact and useful form for later analysis and archival purposes.

Werth, Aaron↗

Cyberattack Detection and Mitigation on Central Volt‐VAr Using Circuit Law and Machine Learning

ABSTRACT In a distribution grid, voltage is maintained within a nominal range through a Volt‐VAr function that controls capacitor banks, reactive power of distributed energy resources (DER), and on‐load tap changers (OLTC). Availability of communications helps with the implementation of central Volt‐VAr control; however, it also opens the system to cyberattacks, causing voltage disturbances. Previous work has shown the adverse impacts of false data injection (FDI) on the central Volt‐VAr control; however, very few works have studied methods to detect and mitigate FDI on Volt‐VAr control. This paper addresses gaps in the detection and mitigation of FDI on the measurement packets of a central Volt‐VAr control. This work uses a two‐stage algorithm for cyberattack detection since the accuracy of a single‐stage machine learning (ML)–based detection method decreases while dealing with unseen data. The first stage is based on the verification of measurements against circuit laws, and the second stage utilizes a tree search algorithm and an ML method to detect the falsified data. This paper compares long short‐term memory (LSTM) and bidirectional LSTM (BiLSTM) as the employed ML algorithms. Finally, the mitigation algorithm replaces the falsified data with the estimated output of the ML algorithm. The effectiveness of the proposed method is tested for several cases using the IEEE 13‐bus test system in PSCAD software.

Beikbabaei, Milad [Bradley Department of Electrica↗

Operator learning for predicting multiscale bubble growth dynamics

We report simulating and predicting multiscale problems that couple multiple physics and dynamics across many orders of spatiotemporal scales is a great challenge that has not been investigated systematically by deep neural networks (DNNs). Herein, we develop a framework based on operator regression, the so-called deep operator network (DeepONet), with the long-term objective to simplify multiscale modeling by avoiding the fragile and time-consuming “hand-shaking” interface algorithms for stitching together heterogeneous descriptions of multiscale phenomena. To this end, as a first step, we investigate if a DeepONet can learn the dynamics of different scale regimes, one at the deterministic macroscale and the other at the stochastic microscale regime with inherent thermal fluctuations. Specifically, we test the effectiveness and accuracy of the DeepONet in predicting multirate bubble growth dynamics, which is described by a Rayleigh–Plesset (R–P) equation at the macroscale and modeled as a stochastic nucleation and cavitation process at the microscale by dissipative particle dynamics (DPD). First, we generate data using the R–P equation for multirate bubble growth dynamics caused by randomly time-varying liquid pressures drawn from Gaussian random fields (GRFs). Our results show that properly trained DeepONets can accurately predict the macroscale bubble growth dynamics and can outperform long short-term memory networks. We also demonstrate that the DeepONet can extrapolate accurately outside the input distribution using only very few new measurements. Subsequently, we train the DeepONet with DPD data corresponding to stochastic bubble growth dynamics. Although the DPD data are noisy and we only collect sparse data points on the trajectories, the trained DeepONet model is able to predict accurately the mean bubble dynamics for time-varying GRF pressures. Taken together, our findings demonstrate that DeepONets can be employed to unify the macroscale and microscale models of the multirate bubble growth problem, hence providing new insight into the role of operator regression via DNNs in tackling realistic multiscale problems and in simplifying modeling with heterogeneous descriptions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Implementing Connected Component Labeling as a User Defined Operator for SciDB

We have implemented a flexible User Defined Operator (UDO) for labeling connected components of a binary mask expressed as an array in SciDB, a parallel distributed database management system based on the array data model. This UDO is able to process very large multidimensional arrays by exploiting SciDB's memory management mechanism that efficiently manipulates arrays whose memory requirements far exceed available physical memory. The UDO takes as primary inputs a binary mask array and a binary stencil array that specifies the connectivity of a given cell to its neighbors. The UDO returns an array of the same shape as the input mask array with each foreground cell containing the label of the component it belongs to. By default, dimensions are treated as non-periodic, but the UDO also accepts optional input parameters to specify periodicity in any of the array dimensions. The UDO requires four stages to completely label connected components. In the first stage, labels are computed for each subarray or chunk of the mask array in parallel across SciDB instances using the weighted quick union (WQU) with half-path compression algorithm. In the second stage, labels around chunk boundaries from the first stage are stored in a temporary SciDB array that is then replicated across all SciDB instances. Equivalences are resolved by again applying the WQU algorithm to these boundary labels. In the third stage, relabeling is done for each chunk using the resolved equivalences. In the fourth stage, the resolved labels, which so far are "flattened" coordinates of the original binary mask array, are renamed with sequential integers for legibility. The UDO is demonstrated on a 3-D mask of O(1011) elements, with O(108) foreground cells and O(106) connected components. The operator completes in 19 minutes using 84 SciDB instances.

UDO↗

MINING NUCLEAR TRANSIENT DATA THROUGH SYMBOLIC CONVERSION

Dynamic Probabilistic Risk Assessment (DPRA) methodologies generate enormous amounts of data for a very large number of simulations. The data contain temporal information of both the state variables of the simulator and the temporal status of specific systems/components. In order to measure system performances, limitations and resilience, such data need to be carefully analyzed with the objective of discovering the correlations between sequence/timing of events and system dynamics. A first approach toward discovering these correlations from data generated by DPRA methodologies has been performed by organizing scenarios into groups using classification or clustering based algorithms. The identification of the correlations between system dynamics and timing/sequencing of events is performed by observing the temporal distribution of these events in each group of scenarios. Instead of performing “a posteriori” analysis of these correlations, this paper shows how it is possible to identify the correlations implicitly by performing a symbolic conversion of both continuous (temporal profiles of simulator state variables) and discrete (status of systems and components) data. Symbolic conversion is performed for each simulation by properly quantizing both continuous and discrete data and then converting them as a series of symbols. After merging both series together, a temporal phrase is obtained. This phrase preserves duration, coincidence and sequence of both continuous and discrete data in a uniform and consistent manner. In this paper it is also shown that by using specific distance measures, it is still possible to post-process such symbolic data using clustering and classification techniques but in considerably less time since the memory needed to store the data is greatly reduced by the symbolic conversion.

97 MATHEMATICS AND COMPUTING↗

Software Synthesis for High Productivity Exascale Computing

Over the three years of our project, we accomplished three key milestones: We demonstrated how ideas from generative programming and software synthesis can help support the development of bulk-synchronous distributed memory kernels. These ideas are realized in a new language called MSL, a C-like language that combines synthesis features with high level notations for array manipulation and bulk-synchronous parallelism to simplify the semantic analysis required for synthesis. We also demonstrated that these high level notations map easily to low level C code and show that the performance of this generated code matches that of handwritten Fortran. Second, we introduced the idea of solver-aided domain-specific languages (SDSLs), which are an emerging class of computer-aided programming systems. SDSLs ease the construction of programs by automating tasks such as verification, debugging, synthesis, and non-deterministic execution. SDSLs are implemented by translating the DSL program into logical constraints. Next, we developed a symbolic virtual machine called Rosette, which simplifies the construction of such SDSLs and their compilers. We have used Rosette to build SynthCL, a subset of OpenCL that supports synthesis. Third, we developed novel numeric algorithms that move as little data as possible, either between levels of a memory hierarchy or between parallel processors over a network. We achieved progress in three aspects of this problem. First we determined lower bounds on communication. Second, we compared these lower bounds to widely used versions of these algorithms, and noted that these widely used algorithms usually communicate asymptotically more than is necessary. Third, we identified or invented new algorithms for most linear algebra problems that do attain these lower bounds, and demonstrated large speed-ups in theory and practice.

97 MATHEMATICS AND COMPUTING↗

Review and analysis of dense linear system solver package for distributed memory machines

A dense linear system solver package recently developed at the University of Texas at Austin for distributed memory machine (e.g. Intel Paragon) has been reviewed and analyzed. The package contains about 45 software routines, some written in FORTRAN, and some in C-language, and forms the basis for parallel/distributed solutions of systems of linear equations encountered in many problems of scientific and engineering nature. The package, being studied by the Computer Applications Branch of the Analysis and Computation Division, may provide a significant computational resource for NASA scientists and engineers in parallel/distributed computing. Since the package is new and not well tested or documented, many of its underlying concepts and implementations were unclear; our task was to review, analyze, and critique the package as a step in the process that will enable scientists and engineers to apply it to the solution of their problems. All routines in the package were reviewed and analyzed. Underlying theory or concepts which exist in the form of published papers or technical reports, or memos, were either obtained from the author, or from the scientific literature; and general algorithms, explanations, examples, and critiques have been provided to explain the workings of these programs. Wherever the things were still unclear, communications were made with the developer (author), either by telephone or by electronic mail, to understand the workings of the routines. Whenever possible, tests were made to verify the concepts and logic employed in their implementations. A detailed report is being separately documented to explain the workings of these routines.

Narang, H. N.↗

ATDM LANL FleCSI: Topology and Execution Framework

FleCSI is a compile-time configurable C++ framework designed to support multi-physics application development. As such, FleCSI attempts to provide a very general set of infrastructure design patterns that can be specialized and extended to suit the needs of a broad variety of solver and data requirements. This means that FleCSI is potentially useful to many different ECP projects. Current support includes multidimensional mesh topology, mesh geometry, and mesh adjacency information, n-dimensional hashed-tree data structures, graph partitioning interfaces, and dependency closures (to identify data dependencies between distributed-memory address spaces). FleCSI introduces a functional programming model with control, execution, and data abstractions that are consistent with state-of-the-art task-based runtimes such as Legion and Charm++. The model also provides support for fine-grained, data-parallel execution with backend support for runtimes such as OpenMP and C++17. The FleCSI abstraction layer provides the developer with insulation from the underlying runtimes, while allowing support for multiple runtime systems, including conventional models like asynchronous MPI. The intent is to give developers a concrete set of user-friendly programming tools that can be used now, while allowing flexibility in choosing runtime implementations and optimizations that can be applied to architectures and runtimes that arise in the future. This project is essential to the ECP Ristra Next-Generation Code project, part of ASC ATDM, because it provides a hierarchically parallel programming model that is consistent with the design of modern system architectures, but which allows for the straightforward expression of algorithmic parallelism in a portably performant manner.

97 MATHEMATICS AND COMPUTING↗

Development and Operation of a Material Identification and Discrimination Imaging Spectroradiometer

Many imaging applications require quantitative determination of a scene's spectral radiance. This paper describes a new system capable of real-time spectroradiometric imagery. Operating at a full-spectrum update rate of 30Hz, this imager is capable of collecting a 30 point spectrum from each of three imaging heads: the first operates from 400 nm to 950 nm, with a 2% bandwidth; the second operates from 1.5 micro-m to 5.5 micro-m with a 1.5% bandwidth; the third operates from 5 micro-m to 12 micro-m, also at a 1.5% bandwidth. Standard image format is 256 x 256, with 512 x 512 possible in the VIS/NIR head. Spectra of up to 256 points are available at proportionately lower frame rates. In order to make such a tremendous amount of data more manageable, internal processing electronics perform four important operations on the spectral imagery data in real-time. First, all data in the spatial/spectral cube of data is spectro-radiometrically calibrated as it is collected. Second, to allow the imager to simulate sensors with arbitrary spectral response, any set of three spectral response functions may be loaded into the imager including delta functions to allow single wavelength viewing; the instrument then evaluates the integral of the product of the scene spectral radiances and the response function. Third, more powerful exploitation of the gathered spectral radiances can be effected by application of various spectral-matched filtering algorithms to identify pixels whose relative spectral radiance distribution matches a sought-after spectral radiance distribution, allowing materials-based identification and discrimination. Fourth, the instrument allows determination of spectral reflectance, surface temperature, and spectral emissivity, also in real-time. The spectral imaging technique used in the instrument allows tailoring of the frame rate and/or the spectral bandwidth to suit the scene radiance levels, i.e., frame rate can be reduced, or bandwidth increased to improve SNR when viewing low radiance scenes. The unique challenges of design and calibration are described. Pixel readout rates of 160 MHz, for full frame readout rates of 1000 Hz (512 x 512 image) present the first challenge; processing rates of nearly 600 million integer operations per second for sensor emulation, or over 2 billion per second for matched filtering, present the second. Spatial and spectral calibration of 66,536 pixels (262,144 for the 512 x 512 version) and up to 1,000 spectral positions mandate novel decoupling methods to keep the required calibration memory to a reasonable size. Large radiometric dynamic range also requires care to maintain precision operation with minimum memory size.

Dombrowski, Mark↗