Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “expanded analysis”

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 1,333 records · Page 74
All sourcesNASA NTRSDOE OSTI
NASA NTRS2007

Inflatable Aerocapture Decelerators for Mars Orbiters

A multi-disciplinary research program was recently completed, sponsored by NASA Marshall Space Flight Center, on the subject of aerocapture of spacecraft weighing up to 5 metric tons at Mars. Heavier spacecraft will require deployable drag area beyond the dimensional limits of current and planned launch fairings. This research focuses on the approach of lightweight inflatable decelerators constructed with thin films, using fiber reinforcement and having a temperature limitation of 500 C. Trajectory analysis defines trajectories for a range of low ballistic coefficients for which convective heat flux is compatible with the material set. Fluid-Structure Interaction (FSI) tools are expanded to include the rarified flow regime. Several non-symmetrical configurations are evaluated for their capability to develop lift as part of the necessary trajectory control strategy. Manufacturing technology is developed for 3-D stretch forming of polyimide films and for tailored fiber reinforcement of thin films. Finally, the mass of the decelerator is estimated and compared to the mass of a traditional rigid aeroshell.

Brown, Glen J.↗
NASA NTRS2007

CECE: A Deep Throttling Demonstrator Cryogenic Engine for NASA's Lunar Lander

As one of the first technology development programs awarded under NASA's Vision for Space Exploration, the Pratt & Whitney Rocketdyne (PWR) Deep Throttling, Common Extensible Cryogenic Engine (CECE) program was selected by NASA in November 2004 to begin technology development and demonstration toward a deep throttling, cryogenic Lunar Lander engine for use across multiple human and robotic lunar exploration mission segments with extensibility to Mars. The CECE program leverages the maturity and previous investment of a flight-proven hydrogen/oxygen expander cycle engine, the RL10, to develop and demonstrate an unprecedented combination of reliability, safety, durability, throttlability, and restart capabilities in a high-energy, cryogenic engine. NASA Marshall Space Flight Center and NASA Glenn Research Center personnel were integral design and analysis team members throughout the requirements assessment, propellant studies and the deep throttling demonstrator elements of the program. The testbed selected for the initial deep throttling demonstration phase of this program was a minimally modified RL10 engine, allowing for maximum current production engine commonality and extensibility with minimum program cost. In just nine months from technical program start, CECE Demonstrator No. 1 engine testing in April/May 2006 at PWR's E06 test stand successfully demonstrated in excess of 10:1 throttling of the hydrogen/oxygen expander cycle engine. This test provided an early demonstration of a viable, enabling cryogenic propulsion concept with invaluable system-level technology data acquisition toward design and development risk mitigation for both the subsequent CECE Demonstrator No. 2 program and to the future Lunar Lander Design, Development, Test and Evaluation effort.

Giuliano, Victor J.↗
NASA NTRS2010

MCNP Simulations of Measurement of Insulation Compaction in the Cryogenic Rocket Fuel Tanks at Kennedy Space Center by Fast/Thermal Neutron Techniques

MCNP simulations have been run to evaluate the feasibility of using a combination of fast and thermal neutrons as a nondestructive method to measure of the compaction of the perlite insulation in the liquid hydrogen and oxygen cryogenic storage tanks at John F. Kennedy Space Center (KSC). Perlite is a feldspathic volcanic rock made up of the major elements Si, AI, Na, K and 0 along with some water. When heated it expands from four to twenty times its original volume which makes it very useful for thermal insulation. The cryogenic tanks at Kennedy Space Center are spherical with outer diameters of 69-70 feet and lined with a layer of expanded perlite with thicknesses on the order of 120 cm. There is evidence that some of the perlite has compacted over time since the tanks were built 1965, affecting the thermal properties and possibly also the structural integrity of the tanks. With commercially available portable neutron generators it is possible to produce simultaneously fluxes of neutrons in two energy ranges: fast (14 Me V) and thermal (25 me V). The two energy ranges produce complementary information. Fast neutrons produce gamma rays by inelastic scattering, which is sensitive to Fe and O. Thermal neutrons produce gamma rays by prompt gamma neutron activation (PGNA) and this is sensitive to Si, Al, Na, K and H. The compaction of the perlite can be measured by the change in gamma ray signal strength which is proportional to the atomic number densities of the constituent elements. The MCNP simulations were made to determine the magnitude of this change. The tank wall was approximated by a I-dimensional slab geometry with an 11/16" outer carbon steel wall, an inner stainless wall and 120 cm thick perlite zone. Runs were made for cases with expanded perlite, compacted perlite or with various void fractions. Runs were also made to simulate the effect of adding a moderator. Tallies were made for decay-time analysis from t=0 to 10 ms; total detected gamma-rays; detected gamma-rays from thermal neutron reactions d. detected gamma-rays from non-thermal neutron reactions and total detected gamma-rays as a function of depth into the annulus volume. These indicated a number of possible independent metrics of perlite compaction. For example the count rate for perlite elements increased from 3600 to 8500 cps for an increase in perlite density from 6 lbs/lcf to 16.5 lbs/cf. Thus the MCNP simulations have confirmed the feasibility of using neutron methods to map the compaction of perlite in the walls of the cryogenic tanks.

Livingston, R. A.↗
NASA NTRS2008

Ares V: Application to Solar System Scientific Exploration

Ares V would revolutionize the way we accomplish solar system exploration -Shift to a better balance of field data acquisition and laboratory analysis -Greatly improve field data acquisition Better instruments, better instrument complements. Larger data volumes returned. Would require a new fiscal paradigm. "Ares V Solar System Science" workshop helped expand the set of possibly-enabled missions.

Elliott, John O.↗
NASA NTRSDate not supplied

Expanding Our Understanding of What Constitutes a Safety-Relevant Occurrence

Focusing on undesired operator behaviors is pervasive in system design and safety management cultures. This focus limits the data that are collected, the questions that are asked during data analysis, and therefore our understanding of what operators do in everyday work. Human performance represents a significant source of safety data that includes both desired and undesired actions. When safety is characterized only in terms of errors and failures, the vast majority of human impacts on system safety and performance are ignored. The outcomes of safety data analyses dictate what is learned from those data, which in turn informs safety policies and safety-related decision making. When learning opportunities are systematically restricted by focusing only on rare failure events, not only do we learn less (and less often), but we can draw misleading conclusions by relying on a non-representative sample of human performance data. The study of human contributions to safety represents a vast and largely unexplored opportunity to learn. Changes in how we define and think about safety can highlight new opportunities for collection and analysis of safety-relevant data. Developing an integrated safety picture to better inform system design, safety-related decision making and policies depends upon identifying, collecting, and interpreting safety producing behaviors in addition to safety reducing behaviors.

Jon B Holbrook↗
DOE OSTI2020

Deciphering Degradation: Machine Learning on Real-World Performance Data

The solar investment community is in need of up-to-date and data-derived system degradation rates to use in financial models that calculate the risk and expectation of energy revenue. To date, financial models choose degradation rates from past studies that are limited and may not be representative of solar projects in development today. As the solar industry expands in all sectors (residential, commercial, and utility), there is a growing amount of energy generation data available to analyze. With the help of the open source degradation analysis code (RdTools) it is now possible to derive degradation on an ongoing basis and continuously provide up-to-date durability statistics to the investment community. Ongoing and accurate statistics of fielded photovoltaic systems allows financial stakeholders to constrain energy revenue projections, lower financial risk, and increase the bankability of solar. This presentation will provide a mid-term progress update on a 2 year SETO-funded degradation project that involves applying RdTools analysis on kWh Analytics’ industry database. With results from the RdTools analysis in hand, machine learning methods are employed to quantify how much variance of system performance degradation can be explained by predictor variables such as environmental factors, equipment bill of materials characteristics, and system design information. The goal of this presentation is to promote the usage of RdTools on an ongoing to data owners and to solicit feedback from researchers and stakeholders on how to maximize impact of the SETO-funded kWh Analytics degradation project.

14 SOLAR ENERGY↗
DOE OSTI2023

LA100 Equity Strategies. Chapter 12: Distribution Grid Upgrades for Equitable Resilience and Solar, Storage, and Electric Vehicle Access

The LA100 Equity Strategies project integrates community guidance with robust research, modeling, and analysis to identify strategy options that can increase equitable outcomes in Los Angeles' clean energy transition. As Los Angeles transitions toward clean energy, existing distribution grid infrastructure will need to be updated and expanded to support reliable service during routine operations, enable interconnection with distributed energy resources and electrified loads, and provide access to energy-related services during disasters. This chapter focuses on equity in distribution grid upgrades, reliability, and resilience in Los Angeles. Specifically, NREL performed grid upgrade and resilience analyses using a detailed model of the distribution grid and income-differentiated household load profiles, electric vehicle (EV) adoption patterns, distributed solar adoption, and grid reliability to explore two key questions to inform how the City of Los Angeles can ensure a resilient and reliable distribution grid for all communities during the clean energy transition: Where can distribution system upgrades can be prioritized to enable equitable access to, and adoption of, clean energy technologies and how can Los Angeles provide equitable, resilient access to electricity-related services (e.g., health care, food) during disaster events like earthquakes and flooding? The electric distribution system is the "last mile" of the grid, linking the multistate bulk power system with customers; new loads, including EVs; and distributed energy resources, such as customer and community solar and storage. This analysis focuses on the 4.8-kilovolt (kV) system, including service transformers that represent the utility-side of the grid connection for most residential customers. Chapter 17 looks at the customer-side of the grid connection with a focus on electric panel upgrade needs. The transition toward clean energy can put additional stress on the distribution system from distributed energy resources and electrification - especially EVs and increased use of electricity for heating, cooling, cooking, and hot water. This stress, measured here as the number of equipment overloads and voltage violations, correlates strongly to grid reliability and therefore is used as a proxy for understanding additional upgrades needed and to help ensure equitable access to electrification and distributed energy resources. NREL also conducted community resilience analysis to examine customer-level access to both electricity and a larger range of services, such as hospitals and grocery stores during a disaster. This analysis explicitly considers equity to understand differences in current resilience and resilience strategies to effectively improve critical services access for all Angelenos. Research was guided by input from the community engagement process, and associated equity strategies are presented in alignment with that guidance.

14 SOLAR ENERGY↗
NASA NTRSDate not supplied

Evidence for Partially Chloritized Smectite in Gale Crater, Mars, and Implications for Diagenesis

Phyllosilicates are an important mineral group found in ancient (~3.5-4.1 Gyr old) martian terrains because they are a marker of water-rock interactions and can help constrain aqueous conditions (e.g., pH, salinity, temperature) and, thus, identify habitable environments. Orbital visible/short-wave infrared (VSWIR) data demonstrate that smectite is the most abundant type of phyllosilicate on Mars, followed by chlorite [e.g., 1]. The geologic settings in which phyllosilicates are found provide important clues into their formation. Smectite has been identified in abundances of up to ~30 wt.% in early Hesperian-aged fluvial-lacustrine sedimentary rocks in Gale crater using the CheMin X-ray diffractometer on the Mars Science Laboratory Curiosity rover [e.g., 2-6]. CheMin XRD patterns and evolved water measured by the Sample Analysis at Mars (SAM) instrument suite show the structure of smectite changes from trioctahedral Fe(II)-bearing smectite at the base of the section to dioctahedral nontronite and montmorillonite ~400 m up section in the Glen Torridon valley where orbital VSWIR show evidence of Fe/Mg smectite. Most of the smectite identified by CheMin is collapsed (i.e., lacking substantial interlayer H2O) based on basal spacings at 10 Å. Fe(II) saponite found in two drill targets at the base of the section in Yellowknife Bay, however, suggest the smectite is expanded. The XRD pattern of the “Cumberland” drill target has a peak at 13.5 Å, whereas the “John Klein” drill target has a peak at 10 Å and a shoulder extending to higher d-spacings (Fig. 1) [2]. A possible explanation for this expanded structure is partial chloritization of the interlayer site caused by the precipitation of small domains of brucite-like sheets [e.g., 7,8]. Here, we synthesize smectite with different degrees of chloritization and analyze the products via XRD, evolved gas analysis (EGA), and VSWIR to determine whether chloritized smectite on Mars can be recognized with these techniques.

E B Rampe↗
DOE OSTI2023

Identifying strengths and weaknesses of methods for computational network inference from single-cell RNA-seq data

Single-cell RNA-sequencing (scRNA-seq) offers unparalleled insight into the transcriptional programs of different cellular states by measuring the transcriptome of thousands of individual cells. An emerging problem in the analysis of scRNA-seq is the inference of transcriptional gene regulatory networks and a number of methods with different learning frameworks have been developed to address this problem. Here, we present an expanded benchmarking study of eleven recent network inference methods on seven published scRNA-seq datasets in human, mouse, and yeast considering different types of gold standard networks and evaluation metrics. We evaluate methods based on their computing requirements as well as on their ability to recover the network structure. We find that, while most methods have a modest recovery of experimentally derived interactions based on global metrics such as Area Under the Precision Recall curve, methods are able to capture targets of regulators that are relevant to the system under study. Among the top performing methods that use only expression were SCENIC, PIDC, MERLIN or Correlation. Addition of prior biological knowledge and the estimation of transcription factor activities resulted in the best overall performance with the Inferelator and MERLIN methods that use prior knowledge outperforming methods that use expression alone. We found that imputation for network inference did not improve network inference accuracy and could be detrimental. Comparisons of inferred networks for comparable bulk conditions showed that the networks inferred from scRNA-seq datasets are often better or at par with the networks inferred from bulk datasets. Our analysis should be beneficial in selecting methods for network inference. At the same time, this highlights the need for improved methods and better gold standards for regulatory network inference from scRNAseq datasets.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Calcium fluoride as a dominating matrix for quantitative analysis by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS): A feasibility study

Here, calcium fluoride formed by the reaction between ammonium bifluoride and calcium chloride was investigated as a dominating matrix for quantitative analysis by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). Transformation from a solid sample to the calcium fluoride-based matrix permitted quantitative analysis based on calibration standards made from elemental standards. A low abundance stable calcium isotope, i.e. 44 Ca + , was monitored as the internal standard for quantitative analysis by LA-ICP-MS. Correlation coefficient factors for multiple elements were obtained with values over 0.999. The results for multiple elements in a certified reference material of soil (NIST SRM 2710a) agreed with the certified values in the range of expanded uncertainty, indicating the present method was valid for quantitation of elements in solid samples.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Metagenomics harvested genus-specific single-stranded DNA-annealing proteins improve and expand recombineering in Pseudomonas species

The widespread Pseudomonas genus comprises a collection of related species with remarkable abilities to degrade plastics and polluted wastes and to produce a broad set of valuable compounds, ranging from bulk chemicals to pharmaceuticals. Pseudomonas possess characteristics of tolerance and stress resistance making them valuable hosts for industrial and environmental biotechnology. However, efficient and high-throughput genetic engineering tools have limited metabolic engineering efforts and applications. To improve their genome editing capabilities, we first employed a computational biology workflow to generate a genus-specific library of potential single-stranded DNA-annealing proteins (SSAPs). Assessment of the library was performed in different Pseudomonas using a high-throughput pooled recombinase screen followed by Oxford Nanopore NGS analysis. Among different active variants with variable levels of allelic replacement frequency (ARF), efficient SSAPs were found and characterized for mediating recombineering in the four tested species. New variants yielded higher ARFs than existing ones in Pseudomonas putida and Pseudomonas aeruginosa, and expanded the field of recombineering in Pseudomonas taiwanensisand Pseudomonas fluorescens. These findings will enhance the mutagenesis capabilities of these members of the Pseudomonas genus, increasing the possibilities for biotransformation and enhancing their potential for synthetic biology applications.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2015

LATE-TIME EVOLUTION OF COMPOSITE SUPERNOVA REMNANTS: DEEP CHANDRA OBSERVATIONS AND HYDRODYNAMICAL MODELING OF A CRUSHED PULSAR WIND NEBULA IN SNR G327.1-1.1

In an effort to better understand the evolution of composite supernova remnants (SNRs) and the eventual fate of relativistic particles injected by their pulsars, we present a multifaceted investigation of the interaction between a pulsar wind nebula (PWN) and its host SNR G327.1-1.1. Our 350 Chandra X-ray observations of SNR G327.1-1.1 reveal a highly complex morphology: a cometary structure resembling a bow shock, prong-like features extending into large arcs in the SNR interior, and thermal emission from the SNR shell. Spectral analysis of the non-thermal emission offers clues about the origin of the PWN structures, while enhanced abundances in the PWN region provide evidence for a mixing of supernova ejecta with PWN material. The overall morphology and spectral properties of the SNR suggest that the PWN has undergone an asymmetric interaction with the SNR reverse shock (RS), whichcan occur as a result of a density gradient in the ambient medium and/or a moving pulsar that displaces the PWN from the center of the remnant. We present hydrodynamical simulations of G327.1-1.1 that show that its morphology and evolution can be described by a ∼17,000-year-old composite SNR that expanded into a density gradient with an orientation perpendicular to the pulsar’s motion. We also show that the RS/PWN interaction scenario can reproduce the broadband spectrum of the PWN from radio to γ-ray wavelengths. The analysis and modeling presented in this work have important implications for our general understanding of the structure and evolution of composite SNRs.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

Propagating Mixed Uncertainties in Cyber Attacker Payoffs: Exploration of Two-Phase Monte Carlo Sampling and Probability Bounds Analysis

Securing cyber-systems on a continual basis against a multitude of adverse events is a challenging undertaking. Game-theoretic approaches, that model actions of strategic decision-makers, are increasingly being applied to address cybersecurity resource allocation challenges. Such game-based models account for multiple player actions and represent cyber attacker payoffs mostly as point utility estimates. Since a cyber-attacker’s payoff generation mechanism is largely unknown, appropriate representation and propagation of uncertainty is a critical task. In this paper we expand on prior work and focus on operationalizing the probabilistic uncertainty quantification framework, for a notional cyber system, through: 1) representation of uncertain attacker and system-related modeling variables as probability distributions and mathematical intervals, and 2) exploration of uncertainty propagation techniques including two-phase Monte Carlo sampling and probability bounds analysis.

cyber security↗
NASA NTRS2015

Late-Time Evolution of Composite Supernova Remnants: Deep Chandra Observations and Hydrodynamical Modeling of a Crushed Pulsar Wind Nebula in SNR G327.1-1.1

In an effort to better understand the evolution of composite supernova remnants (SNRs) and the eventual fate of relativistic particles injected by their pulsars, we present a multifaceted investigation of the interaction between a pulsar wind nebula (PWN) and its host SNR G327.1-1.1. Our 350 ks Chandra X-ray observations of SNR G327.1-1.1 reveal a highly complex morphology; a cometary structure resembling a bow shock, prong-like features extending into large arcs in the SNR interior, and thermal emission from the SNR shell. Spectral analysis of the non-thermal emission offers clues about the origin of the PWN structures, while enhanced abundances in the PWN region provide evidence for mixing of supernova ejecta with PWN material. The overall morphology and spectral properties of the SNR suggest that the PWN has undergone an asymmetric interaction with the SNR reverse shock(RS) that can occur as a result of a density gradient in the ambient medium and or a moving pulsar that displaces the PWN from the center of the remnant. We present hydrodynamical simulations of G327.1-1.1 that show that its morphology and evolution can be described by a approx. 17,000 yr old composite SNR that expanded into a density gradient with an orientation perpendicular to the pulsar's motion. We also show that the RSPWN interaction scenario can reproduce the broadband spectrum of the PWN from radio to gamma-ray wavelengths. The analysis and modeling presented in this work have important implications for our general understanding of the structure and evolution of composite SNRs.

Evolution↗
DOE OSTI2024

Investigation of pyruvic acid photolysis at the air-liquid interface as a source of aqueous secondary organic aerosols

Pyruvic acid (PA) is a ubiquitous 2-oxocarboxylic acid in the atmosphere. Its photochemical process at the air-liquid (a-l) interface has been suggested as an important source of aqueous secondary organic aerosols. Here, we investigated the photochemical reaction pathways of PA at the a-l interface using synchrotron-based vacuum ultraviolet single-photon ionization mass spectrometry (VUV SPI-MS) coupled with the System for Analysis at the Liquid Vacuum Interface (SALVI) microreactor. Results from mass spectral analysis and the determination of appearance energies (AEs) indicate that photolysis of PA can generate radicals, then they recombine with carboxylic acids and simple molecular oligomers. Furthermore, the preliminary products could form larger oligomers via radical reaction or esterification in the presence of hydroxyl and carboxyl functional groups. Mass spectral comparison shows that most photochemical reactions would complete within 4 h. The expanded photochemistry-driven reaction flowchart of PA is proposed based on the newly discovered products. Our results reveal that the interfacial PA photochemical reactions have different mechanisms from the bulk liquid due to the interfacial properties, such as molecular density, composition, and ion concentration. Our findings show that in situ mass spectral analysis with bright photon ionization is useful to elucidate the contribution of a-l interfacial reactions leading to aqSOA formation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Analysis of ultrafast magnetization switching dynamics in exchange-coupled ferromagnet–ferrimagnet heterostructures

Magnetization switching in ferromagnets has so far been limited to the current-induced spin–orbit–torque effects. Recent observation of helicity-independent all-optical magnetization switching (HI-AOS) in an exchange-coupled ferromagnet–ferrimagnet (FM-FEM) heterostructures expanded the range and applicability of such ultrafast heat-driven magnetization switching. Here we report the element-resolved HI-AOS dynamics of such an exchange-coupled system, using a modified microscopic three-temperature model. We have studied the effect of (i) the Curie temperature of the FM, (ii) FEM composition, (iii) the long-range Ruderman–Kittel–Kasuya–Yosida (RKKY) exchange-coupling strength, and (iv) the absorbed optical energy on the element-specific time-resolved magnetization dynamics. The phase-space of magnetization illustrates how the RKKY coupling strength and the absorbed optical energy influence the switching time. Our analysis demonstrates that the threshold switching energy depends on the composition of the FEM and the switching time depends on the Curie temperature of the FM as well as RKKY coupling strength. Importantly, this simulation anticipates new insights into developing faster and more energy-efficient spintronics devices.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Comparative transcriptome analysis of cells from different areas reveals ROS responsive mechanism at sclerotial initiation stage in Morchella importuna

Morels are some of the most highly prized edible and medicinal mushrooms, with great economic and scientific value. Outdoor cultivation has been achieved and expanded on a large scale in China in recent years. Sclerotial formation is one of the most important phases during the morel life cycle, and previous reports indicated that reactive oxygen species (ROS) play an important role. However, ROS response mechanisms at sclerotial initiation (SI) stage are poorly understood. In this study, comparative transcriptome analyses were performed with sclerotial and hyphal cells at different areas in the same plate at SI stage. Gene expression was significantly different at SI stage between sclerotial formation and mycelia growth areas. GO and KEGG analyses indicated more vigorous metabolic characteristics in the hyphae area, while transcription process, DNA repair, and protein processing were enriched in sclerotial cells. Gene expression related to H 2 O 2 production was high in the hyphae area, while expression of H 2 O 2 -scavenging genes was high in sclerotial cells, leading to a higher H 2 O 2 concentration in the hyphal region than in the sclerotium. Minor differences were observed in gene expression of H 2 O 2 -induced signaling pathway in sclerotial and hyphal cells; however, expression levels of the target genes of transcription factor MSN2, important in the H 2 O 2 -induced signaling pathways, were significantly different. MSN2 enhanced stress response regulation in sclerotia by regulating these target genes. Small molecular HSPs were also found upregulated in sclerotial cells. This study indicated that sclerotial cells are more resistant to ROS stress than hyphal cells through transcriptional regulation of related genes.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Exploring the benefits of using co-packaged optics in data center and AI supercomputer networks: a simulation-based analysis [Invited]

We investigate the advantages of using co-packaged optics in next-generation data center and AI supercomputer networks. The increased escape bandwidth offered by co-packaged optics provides multiple possibilities for building 50T switches and beyond, expanding the opportunities in both the data center and supercomputing domains. Furthermore, this provides network architects with the opportunity to expand their design space and develop simplified networks with enhanced network locality properties. Co-packaging at the switch and server points enables networks with double capacity while reducing the switch count by 64% compared to state-of-the-art systems. We evaluate these concepts through discrete-event simulations using all-to-all and all-reduce traffic patterns that simulate collective communications commonly found in network-bound applications. Initially, we investigate the all-to-all overhead involved in distributing the virtual machines of the applications across multiple leaf switches and compare it to the scenario in which all VMs are placed under a single switch. Subsequently, we evaluate the performance of an AI supercomputing cluster by simulating both patterns for different message sizes, while also varying the number of participating nodes. The results suggest that networks with improved locality properties become increasingly important as the network stack operates at higher speeds; for a stack latency of 1.25 µs, placing the applications under multiple switches can result in up to 68% higher completion times than placing them under a single switch. For AI supercomputers, significant improvements are observed in the mean server throughput, reaching more than 90% for configurations involving 256 nodes and message sizes of at least 128 KiB.

99 GENERAL AND MISCELLANEOUS↗