Search NASA⌕ Search

SEARCH · Search NASA

Results for “node usage”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Introducing the 9500 Node Distribution Test System to Support Advanced Power Applications: An Operations-Focused Approach

The 9500 Node Test System is a representative section of distribution power system model developed as a part of the GridAPPS-D™ project, an effort funded by DOE as a part of the Grid Modernization Lab Consortium (GMLC) program. The test system was developed to fulfill a growing need to represent the rapidly evolving state of electric distribution systems by combining elements of legacy infrastructure systems, modern feeder topologies, and an anticipated future with smart grid technologies. It also provides a network model capable of supporting the simulation of operational scenarios such as the ones in a utility distribution control center. This test system allows the evaluation of the performance of advanced power applications in real-time, such as one that simulates the operations of an Advanced Distribution Management Systems (ADMS), Distributed Energy Resource Management Systems (DERMS), etc. in a Distribution control center. This model is an extension of the widely used IEEE 8500 Node Test Feeder and is currently being validated to become an IEEE test case to help increase adoption and widespread usage among both academia and industry. It is a full-size model representative of a section of a utility’s distribution system with multiple feeders fed from different substations. The model includes multiple distribution circuits, a sub-transmission system, multiple substations, behind the meter customer rooftop photovoltaics (PV), and multiple utility-scale distributed energy resources. To enable accurate simulations of operational scenarios, the 9500 Node Test System is designed to support procedure-based operations, with the ability to realistically demonstrate switching operations, feeder reconfiguration, adjustment of volt-var control equipment, dispatch of distributed generation, and response to planned and unplanned outages. The 9500 Node Test System includes three radial distribution feeders with 12.3 MVA of average load, consisting of both medium voltage and low voltage equipment each supplied by a different distribution substation. The three distribution feeders are connected to each other through Normally-Open switches which can be closed when needed to simulate restoration scenarios due to a fault. One feeder represents today’s grid with low penetration of customer-side renewables. The second represents a potential future grid with microgrids and 100% renewable penetration. The third has no customer generation resources, a district steam plant, and a utility-scale solar farm. The three diverse circuits were created to allow the simulation of both today’s situation as well as potential future scenarios. All three feeders have customers connected by low-voltage secondary triplex lines. This test system meets all requirements outlined in the report for creating a simulation environment that would enable discussion between key technical stakeholders as well as having the potential to accelerate operational application development and their subsequent testing and integration. The new model is a possible representation of what we believe the distribution grid may look like in the future: a high penetration of renewables, reconfigurable radial and mesh topology, numerous DERs, islanded microgrids, and significantly increased data and measurement density. The system supports both the solution of existing and newer algorithms but also enables the evaluation of applications in a realistic operational environment defined by task oriented procedural steps that represent the interaction between the control center operator and field personnel.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2021

A particle-in-cell code comparison for ion acceleration: EPOCH, LSP, and WarpX

There are now more Particle-in-Cell (PIC) codes than ever before that researchers use to simulate intense laser-plasma interactions. To date, there have been relatively few direct comparisons of these codes in the literature, especially for relativistic intensity lasers interacting with thin overdense targets. To address this we perform a code comparison of three PIC codes: EPOCH, LSP, and WarpX for the problem of laser-driven ion acceleration in a 2D(3v) geometry for a 10 20 W cm -2 intensity laser. We examine the plasma density, ion energy spectra, and laser-plasma coupling of the three codes and find strong agreement. We also run the same simulation 20 times with different random seeds to explore statistical fluctuations of the outputs. We then compare the execution times and memory usage of the codes (without “tuning” to improve performance) using between 1 and 48 processors on one node. We provide input files to encourage larger and more frequent code comparisons in this field.

36 MATERIALS SCIENCE↗
NASA NTRS1993

Optimal cube-connected cube multiprocessors

Many CFD (computational fluid dynamics) and other scientific applications can be partitioned into subproblems. However, in general the partitioned subproblems are very large. They demand high performance computing power themselves, and the solutions of the subproblems have to be combined at each time step. The cube-connect cube (CCCube) architecture is studied. The CCCube architecture is an extended hypercube structure with each node represented as a cube. It requires fewer physical links between nodes than the hypercube, and provides the same communication support as the hypercube does on many applications. The reduced physical links can be used to enhance the bandwidth of the remaining links and, therefore, enhance the overall performance. The concept and the method to obtain optimal CCCubes, which are the CCCubes with a minimum number of links under a given total number of nodes, are proposed. The superiority of optimal CCCubes over standard hypercubes was also shown in terms of the link usage in the embedding of a binomial tree. A useful computation structure based on a semi-binomial tree for divide-and-conquer type of parallel algorithms was identified. It was shown that this structure can be implemented in optimal CCCubes without performance degradation compared with regular hypercubes. The result presented should provide a useful approach to design of scientific parallel computers.

Sun, Xian-He↗
NASA NTRSDate not supplied

A New Statistical Estimate of the Radar Coverage of the Low Earth Orbit Debris Environment

For over three decades, the NASA Orbital Debris Program Office (ODPO) has used the Goldstone Orbital Debris Radar, Haystack Ultrawideband Satellite Imaging Radar (HUSIR), and Haystack Auxiliary (HAX) radar assets to collect data on the low Earth orbit (LEO) debris environment. Each radar, with its unique beamwidth, altitude and inclination coverage, and limiting size threshold, operates in a beam park mode to statistically sample the orbital debris population in LEO. Provided that these assets are shared with other users, the orbital debris data collection is not continuous; rather, intermittent data collects are acquired and sent to the NASA ODPO. To understand the sampling process conducted by each radar over time and any related observational biases, a Statistical Confirmation of Radar Uniformity or Bias (SCRUB) code has been implemented to model the coverage of these assets for informing future operations, as well as usage of the data collected from these ground-based sensors. For this analysis, Right Ascension of the Ascending Node (RAAN) is used as a metric to measure statistical coverage. A complete survey of the LEO environment is understood to be measurements that sufficiently sample all values of RAAN for each altitude-inclination pair visible from the radar asset. Regions of incompleteness or statistical bias can help inform future observation campaigns. The SCRUB tool evaluates the coverage of the LEO environment, not by examining individual objects that may pass through a sensor’s field of view (FOV), but by determining which orbit planes pass through the FOV. Once a pointing geometry for a radar site, observation time, and range extent are configured by the user, SCRUB computes the inclinations and altitudes that are visible by the sensor. For each inclination-altitude pair, there is a distinct pair of possible RAAN values, corresponding to the ascending and descending orbit passing through that point in the sensor cone. Repeating this process for all points in the beam, and for multiple time periods during an observation window, creates a matrix of all inclination-altitude-RAAN combinations that are visible during a sensor run. This process can then be repeated for all observations within a year (for an annual survey), propagating all the RAAN values to a common epoch, typically the start of a calendar year, and combining the observations to assemble a full estimate of the RAAN coverage of the LEO environment. This RAAN coverage can then be analyzed for uniformity of sampling, within a certain inclination-altitude pair, or between larger regions of the space environment. This paper provides a general overview of the radar assets utilized by ODPO, typical analysis data products assuming circular orbits, and a discussion of the algorithms that feed between modeling and measurement operations. Following the description of the algorithm, estimates of coverage using HUSIR radar data collected from multiple years are developed and compared.

Chris L Ostrom↗
NASA NTRS1994

The Electronic Documentation Project in the NASA mission control center environment

NASA's space programs like many other technical programs of its magnitude is supported by a large volume of technical documents. These documents are not only diverse but also abundant. Management, maintenance, and retrieval of these documents is a challenging problem by itself; but, relating and cross-referencing this wealth of information when it is all on a medium of paper is an even greater challenge. The Electronic Documentation Project (EDP) is to provide an electronic system capable of developing, distributing and controlling changes for crew/ground controller procedures and related documents. There are two primary motives for the solution. The first motive is to reduce the cost of maintaining the current paper based method of operations by replacing paper documents with electronic information storage and retrieval. And, the other is to improve the efficiency and provide enhanced flexibility in document usage. Initially, the current paper based system will be faithfully reproduced in an electronic format to be used in the document viewing system. In addition, this metaphor will have hypertext extensions. Hypertext features support basic functions such as full text searches, key word searches, data retrieval, and traversal between nodes of information as well as speeding up the data access rate. They enable related but separate documents to have relationships, and allow the user to explore information naturally through non-linear link traversals. The basic operational requirements of the document viewing system are to: provide an electronic corollary to the current method of paper based document usage; supplement and ultimately replace paper-based documents; maintain focused toward control center operations such as Flight Data File, Flight Rules and Console Handbook viewing; and be available NASA wide.

Wang, Lui↗
DOE OSTI2021

The Challenge of Disproportionate Importance of Temporal Features in Predicting HPC Power Consumption

In this work, we demonstrate the challenges in predicting HPC cluster power consumption in the face of significant temporal skew in power consumption behavioral patterns. Predicting large power swings that extend several megawatts has significant operational value for HPC centers, however, prediction is challenging due to the relative rarity of such events and also due to the abrupt or disjoint deviation from the average power consumption levels. To study the impact of this challenge, we have trained a recurrent neural network (RNN) as a reasonably sophisticated model to predict power consumption of the one-year worth of node power consumption data from the Summit supercomputer located in the Oak Ridge Leadership Computing Facility. By studying the prediction results, we have found that although simple usage of RNN models can provide good results on average power consumption levels, it would fail at predicting the power swings that have more operational value. With such results, we discuss potential next steps in addressing such issues aiming towards a robust usage of power prediction techniques in HPC operations.

Li, Chengcheng↗
DOE OSTI2022

Enabling power measurement and control on Astra: The first petascale Arm supercomputer

Astra, deployed in 2018, was the first petascale supercomputer to utilize processors based on the ARM instruction set. The system was also the first under Sandia's Vanguard program which seeks to provide an evaluation vehicle for novel technologies that with refinement could be utilized in demanding, large-scale HPC environments. In addition to ARM, several other important first-of-a-kind developments were used in the machine, including new approaches to cooling the datacenter and machine. Here we document our experiences building a power measurement and control infrastructure for Astra. While this is often beyond the control of users today, the accurate measurement, cataloging, and evaluation of power, as our experiences show, is critical to the successful deployment of a large-scale platform. While such systems exist in part for other architectures, Astra required new development to support the novel Marvell ThunderX2 processor used in compute nodes. In addition to documenting the measurement of power during system bring up and for subsequent on-going routine use, we present results associated with controlling the power usage of the processor, an area which is becoming of progressively greater interest as data centers and supercomputing sites look to improve compute/energy efficiency and find additional sources for full system optimization.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Scalable Pattern Matching in Metadata Graphs via Constraint Checking

Pattern matching is a fundamental tool for answering complex graph queries. Unfortunately, existing solutions have limited capabilities: They do not scale to process large graphs and/or support only a restricted set of search templates or usage scenarios. Moreover, the algorithms at the core of the existing techniques are not suitable for today’s graph processing infrastructures relying on horizontal scalability and shared-nothing clusters, as most of these algorithms are inherently sequential and difficult to parallelize. In this article we present an algorithmic pipeline that bases pattern matching on constraint checking. The key intuition is that each vertex and edge participating in a match has to meet a set of constraints implicitly specified by the search template. These constraints can be verified independently and typically are less expensive to compute than searching the full template. The pipeline we propose generates these constraints and iterates over them to eliminate all the vertices and edges that do not participate in any match, thus reducing the background graph to a subgraph that is the union of all template matches—the complete set of all vertices and edges that participate in at least one match. Additional analysis can be performed on this annotated, reduced graph, such as full match enumeration, match counting, or computing vertex/edge centrality. Furthermore, a vertex-centric formulation for constraint checking algorithms exists, and this makes it possible to harness existing high-performance, vertex-centric graph processing frameworks. This technique (i) enables highly scalable pattern matching in metadata (labeled) graphs; (ii) supports arbitrary patterns with 100% precision; (iii) enables tradeoffs between precision and time-to-solution, while always selects all vertices and edges that participate in matches, thus offering 100% recall; and (iv) supports a set of popular data analytics scenarios. We implement our approach on top of HavoqGT, an open-source asynchronous graph processing framework, and demonstrate its advantages through strong and weak scaling experiments on massive scale real-world (up to 257 billion edges) and synthetic (up to 4.4 trillion edges) labeled graphs, respectively, and at scales (1,024 nodes / 36,864 cores), orders of magnitude larger than used in the past for similar problems. This article serves two purposes: First, it synthesises the knowledge accumulated during a long-term project. Second, it presents new system features, usage scenarios, optimizations, and comparisons with related work that strengthen the confidence that pattern matching based on iterative pruning via constraint checking is an effective and scalable approach in practice. The new contributions include the following: (i) We demonstrate the ability of the constraint checking approach to efficiently support two additional search scenarios that often emerge in practice, interactive incremental search and exploratory search. (ii) We empirically compare our solution with two additional state-of-the-art systems, Arabsque and TriAD. (iii) We show the ability of our solution to accommodate a more diverse range of datasets with varying properties, e.g., scale, skewness, label distribution, and match frequency. (iv) We introduce or extend a number of system features (e.g., work aggregation, load balancing, and the ability to cap the generated traffic) and design optimizations and demonstrate their advantages with respect to improving performance and scalability. (v) We present bottleneck analysis and insights into artifacts that influence performance. (vi) We present a theoretical complexity argument that motivates the performance gains we observe.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2022

Environmental Testing of the OVEN System for Lunar Water Extraction and Prospecting

Introduction: The presence of water ice in permanently shadowed regions on the lunar surface [1] may enable a sustained human presence on the Moon with minimal need for consumables. However, in order to develop a long term utilization plan that includes the usage of in-situ water we must first understand the abundance, stratigraphy and distribution of this re-source. Multiple space agencies currently have plans for lunar water prospecting missions. The Optimized Volatile Extraction Node (OVEN) was designed for water prospecting missions that require samples to be weighed, sealed, and heated as the means of determining water concentration. This method of water quantification necessitates a fair amount of automation, so a rigorous environmental test program was performed in order to build confidence in the performance of the OVEN design. The work presented here describes the OVEN environmental test program as well as ongoing efforts to improve on the design. Vibration: The OVEN participated in two rounds of random vibration tests. The first test was a stand-alone test performed at the Energy Systems Test Area of the Johnson Space Center. The second test was an integrated test with the mobile platform developed for the Resource Prospector project. The OVEN survived both tests without damage, but the tests did provide valuable lessons learned with regards to specific operations. Thermal Vacuum: The OVEN was successfully demonstrated at a temperature range of -50 to 75 C in a thermal vacuum chamber. The need to heat motor gearboxes at lower temperatures was predetermined so this test program was completed by implementing a method of gearbox heating that used the existing circuitry within the motors. Dust: A custom dynamometer was built in order to determine the torque required to move the various mechanisms within the OVEN at a range of temperatures. The OVEN system was coated with lunar dust simulant in order to determine mechanism torques under a worst-case operating condition. Sublimation: Sublimation losses within the OVEN were quantified through a series of test configurations, including an integrated test in a thermal vacuum chamber at Glenn Research Center [2]. Current Work: The OVEN subsystem is currently not a component of any existing prospecting missions, but work continues that will take the lessons learned from previous environmental tests and improve on the design in order to be considered for future prospecting opportunities. References: [1] Colaprete, A., Schultz, P., Heldmann, J., Wooden, D., Shirley, M., Ennico, K., ... & Sollitt, L. (2010). Detection of water in the LCROSS ejecta plume. science, 330(6003), 463-468. [2] Kleinhenz, J., Smith, J., Roush, T., Colaprete, A., Zacny, K., Paulsen, G., ... & Paz, A. (2018). Volatiles Loss from water bearing regolith simulant at Lunar Environments. In Earth and Space 2018: Engineering for Extreme Environments (pp. 454-466). Reston, VA: American Society of Civil Engineers.

A Paz↗
NASA NTRS1978

CAVE3: A general transient heat transfer computer code utilizing eigenvectors and eigenvalues

The method of solution is a hybrid analytical numerical technique which utilizes eigenvalues and eigenvectors. The method is inherently stable, permitting large time steps even with the best of conductors with the finest of mesh sizes which can provide a factor of five reduction in machine time compared to conventional explicit finite difference methods when structures with small time constants are analyzed over long time periods. This code will find utility in analyzing hypersonic missile and aircraft structures which fall naturally into this class. The code is a completely general one in that problems involving any geometry, boundary conditions and materials can be analyzed. This is made possible by requiring the user to establish the thermal network conductances between nodes. Dynamic storage allocation is used to minimize core storage requirements. This report is primarily a user's manual for CAVE3 code. Input and output formats are presented and explained. Sample problems are included which illustrate the usage of the code as well as establish the validity and accuracy of the method.

Palmieri, J. V.↗
DOE OSTI2022

Pseudonymized User-Perspective Summit Login Node Data for 2020 and 2021

This dataset contains hourly snapshot data from each of the 5 login nodes of the Summit supercomputer at Oak Ridge Leadership Computing Facility (OLCF) over a period of 2 years, starting January 2020 and ending after December 2021. The snapshots include lists of currently logged-in users, CPU and memory usage, status of users' batch jobs, and disk usage statistics. Usernames, project identifiers, and file paths have been pseudonymized in order to allow studies of user behavior without divulging Personally Identifiable Information (PII).

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2017

Expanding a Supercomputer Facility Using Modular Data Center Technology

With the expansion of high-end computing resources needed to support NASA's increasing demands for physics-based simulations, the facility housing Pleiades-the agency's largest supercomputer-recently reached its power and cooling capacity. In response, the NASA Advanced Supercomputing Division at Ames Research Center undertook a prototype project that resulted in a new facility based on modular data center technology. The facility, a ~1000 square-foot module on a concrete pad with room for 16-18 compute racks, was completed in fall 2016 and an SGI computer system, named Electra, was deployed there in early 2017. Cooling is performed via an evaporative system built into the module, and preliminary experience shows a Power Usage Effectiveness (PUE) of ~1.03. Electra achieved over a petaflop on the LINPACK benchmark, sufficient to rank number 96 on the November 2016 TOP500 list. The system consists of 1,152 InfiniBand-connected Intel Xeon Broadwell-based nodes. Its users access their files on a facility wide file system shared by all compute assets via Mellanox MetroX InfiniBand extenders, which connect the Electra fabric to Lustre routers InfiniBand fabric over fiber-optic links about 300 meters long. The prototype has exceeded expectations and is serving as a blueprint for future expansions.*†

Facility↗
NASA NTRS2019

Electra: A Modular-Based Expansion of NASA's Supercomputing Capability

NASA has increasingly relied on high-performance computing (HPC) re- sources for computational modeling, simulation, and data analysis to meet the science and engineering goals of its missions in space exploration, aeronautics, and Earth and space science. The NASA Advanced Supercomputing (NAS) Division at Ames Research Center in Silicon Valley, Calif., hosts NASA’s premier supercomputing resources, integral to achieving and enhancing the success of the agency’s missions. NAS provides a balanced environment, funded under the High-End Computing Capability (HECC) project, comprised of world-class supercomputers, including its flagship distributed-memory cluster, Pleiades; high-speed networking; and massive data storage facilities, along with multi-disciplinary support teams for user support, code porting and optimization, and large-scale data analysis and scientific visualization. However, as scientists have increased the fidelity of their simulations and engineers are conducting larger parameter-space studies, the requirements for supercomputing resources have been growing by leaps and bounds. With the facility housing the HECC systems reaching its power and cooling capacity, NAS undertook a prototype project to investigate an alternative approach for housing supercomputers. Modular supercomputing, or container-based computing, is an innovative concept for expanding NASA’s HPC capabilities. With modular supercomputing, additional containers—similar to portable storage pods—can be connected together as needed to accommodate the agency’s ever-increasing demand for computing resources. In addition, taking advantage of the local weather permits the use of cooling technologies that would additionally save energy and reduce annual water usage. The first stage of NASA’s Modular Supercomputing Facility (MSF) prototype, which resulted in a 1,000 square-foot module on a concrete pad with room for 16 compute racks, was completed in Fall 2016 and an SGI (now HPE) computer system, named Electra, was deployed there in early 2017. Cooling is performed via an evaporative system built into the module, and preliminary experience shows a Power Usage Effectiveness (PUE) measurement of 1.03. Electra achieved over a petaflop on the LINPACK benchmark, sufficient to rank number 96 on the November 2016 TOP500 list [14]. The system consists of 1,152 InfiniBand-connected Intel Xeon Broadwell-based nodes. Its users access their files on a facility-wide file system shared by all HECC compute assets via Mellanox MetroX InfiniBand extenders, which connect the Electra fabric to Lustre routers in the primary facility over fiber-optic links about 900 feet long. The MSF prototype has exceeded expectations and is serving as a blueprint for future expansions. In the remainder of this chapter, we detail how modular data center technology can be used to expand an existing compute resource. We begin by describing NASA’s requirements for supercomputing and how resources were provided prior to the integration of the Electra module-based system.

Biswas, Rupak↗
DOE OSTI2026

Integral Kernel Methods for Nonlinear Parabolic-Elliptic Systems

Nonlinear parabolic-elliptic systems arise in many physical, biological, and chemical phenomena such as chemotaxis, ion transport, self-gravitating particles, and Brownian vortices. Existing methods struggle with the strong coupling and high nonlinearity and nonlocality of some of these systems, especially the ill-conditioned, convection-dominated problems. To overcome numerical difficulties, current approaches rely on initial guesses, preconditioning, or iterative techniques with no convergence guarantees. They might suffer from poor scalability, large memory usage, and difficulty to parallelize. Inspired by the connection of parabolic-elliptic systems to stochastic processes, we introduce a novel meshless, monolithic, and fully explicit method that naturally encapsulates the elliptic and parabolic operators into a single step which updates each node deterministically with global information. By being fully quadrature-based, it avoids solving systems of discretized equations and does not utilize initial guesses or preconditioning, while requiring little memory and being easy to parallelize. We first derive the method in an integral kernel formulation with quadratic complexity in the number of integration nodes and then leverage kernel-independent fast multipole methods (FMM) to present a scalable algorithm with linear complexity. We provide numerical examples for the Poisson-Nernst-Planck equations in one, two, and three dimensions, together with the derivation of the integral kernel for each case. Furthermore, the examples demonstrate the fast convergence and scalability of the FMM-accelerated algorithm, as well as its suitability for convection-dominated problems, making it competitive against traditional PDE solvers.

PDE systems↗
NASA NTRS1991

Evaluation of Fungal Metabolic Compounds Released to the Air in a Restricted Environment

The metabolic action of selected fungi species on common components of the interior of Space Station Freedom (SSF) will be tested. When present, volatile organic chemicals will be collected on porous polymer adsorbent columns. Using thermal desorption, the volatile compounds will be passed onto a gas chromatographic column for analysis. The Space Station Freedom (SSF) modular complex will largely be individually self contained and the established air environment will not be easily adjusted. The development and maintenance of a safe working environment offers a considerable challenge. Present plans for use of SSF acknowledge periods of manned activities and alternate times when the station is unmanned. The obvious necessity for clean and safe air and water during periods of use have been pursued as fundamental systems to SSF success. Somewhat less obvious, although perhaps of no less importance to the success of long term cyclic usage, are those periods of inactivity. It is during these periods when spores from microorganisms may be afforded the best conditions to germinate and in the vegetative form react with the complex synthetic chemical polymers which compose the furnishings and hardware of SSF nodes. Biodegradation could constitute a real hygiene problem, if the organisms form and release volatile organic chemicals. Similar problems have been documented in closed and improperly ventilated buildings and work spaces. Many of the metabolic products of fungi and bacterial growth create a variety of health problems. Analytical chemical techniques will first be used to document the growth of Aspergillus, Penicillium, and Cladosporium fungal species on the potential substrates Nomex and Kevlar. Any volatile organics that are released will be measured using the spectrum of gas adsorption chromatography. The level of microbial contamination that is necessary to produce such volatile compounds and the relative amounts expected to accumulate will be estimated.

Ferebee, Robert N.↗
DOE OSTI2020

Optimizing High Performance Markov Clustering for Pre-Exascale Architectures

HipMCL is a high-performance distributed memory implementation of the popular Markov Cluster Algorithm (MCL) and can cluster large-scale networks within hours using a few thousand CPU-equipped nodes. It relies on sparse matrix computations and heavily makes use of the sparse matrix-sparse matrix multiplication kernel (SpGEMM). The existing parallel algorithms in HipMCL are not scalable to Exascale architectures, both due to their communication costs dominating the runtime at large concurrencies and also due to their inability to take advantage of accelerators that are increasingly popular. In this work, we systematically remove scalability and performance bottlenecks of HipMCL. We enable GPUs by performing the expensive expansion phase of the MCL algorithm on GPU. Additionally, we propose a CPU-GPU joint distributed SpGEMM algorithm called pipelined Sparse SUMMA and integrate a probabilistic memory requirement estimator that is fast and accurate. Furthermore, we develop a new merging algorithm for the incremental processing of partial results produced by the GPUs, which improves the overlap efficiency and the peak memory usage. We also integrate a recent and faster algorithm for performing SpGEMM on CPUs. We validate our new algorithms and optimizations with extensive evaluations. With the enabling of the GPUs and integration of new algorithms, HipMCL is up to 12.4x faster, being able to cluster a network with 70 million proteins and 68 billion connections just under 15 minutes using 1024 nodes of ORNL's Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Designing and prototyping extensions to the Message Passing Interface in MPICH

As HPC system architectures and the applications running on them continue to evolve, the MPI standard itself must evolve. The trend in current and future HPC systems toward powerful nodes with multiple CPU cores and multiple GPU accelerators makes efficient support for hybrid programming critical for applications to achieve high performance. However, the support for hybrid programming in the MPI standard has not kept up with recent trends. The MPICH implementation of MPI provides a platform for implementing and experimenting with new proposals and extensions to fill this gap and to gain valuable experience and feedback before the MPI Forum can consider them for standardization. Here, in this work, we detail six extensions implemented in MPICH to increase MPI interoperability with other runtimes, with a specific focus on heterogeneous architectures. First, the extension to MPI generalized requests lets applications integrate asynchronous tasks into MPI’s progress engine. Second, the iovec extension to datatypes lets applications use MPI datatypes as a general-purpose data layout API beyond just MPI communications. Third, a new MPI object, MPIX_Stream, can be used by applications to identify execution contexts beyond MPI processes, including threads and GPU streams. MPIX stream communicators can be created to make existing MPI functions thread-aware and GPU-aware, thus providing applications with explicit ways to achieve higher performance. Fourth, MPIX Streams are extended to support the enqueue semantics for offloading MPI communications onto a GPU stream context. Fifth, thread communicators allow MPI communicators to be constructed with individual threads, thus providing a new level of interoperability between MPI and on-node runtimes such as OpenMP. Lastly, we present an extension to invoke MPI progress, which lets users spawn progress threads with fine-grained control to adapt the communication performance to their application designs. We describe the design and implementation of these extensions, provide usage examples, and highlight their expected benefits with performance results.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2002

Distributed topology control algorithm for multihop wireless netoworks

We present a network initialization algorithmfor wireless networks with distributed intelligence. Each node (agent) has only local, incomplete knowledge and it must make local decisions to meet a predefined global objective. Our objective is to use power control to establish a topology based onthe relative neighborhood graph which has good overall performance in terms of power usage, low interference, and reliability.

topology control distributed algorithm wireless ne↗