Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “bottleneck 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 415 records · Page 23
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Heterostructured Au–Ir Catalysts for Enhanced Oxygen Evolution Reaction

Electrochemical water splitting operated under acidic conditions provides a clean approach to generate hydrogen fuels. Currently, the sluggish kinetics of oxygen evolution reaction (OER) at the anode is a bottleneck limiting the acidic water splitting. Here, we report that the OER activity of Ir, one of the most commonly used OER catalysts, can be boosted by forming phase boundaries with Au. Mixed Au and Ir catalysts were synthesized on carbon paper electrodes, which underwent structural evolution under the OER environment to form an Au-Ir interface-rich structure. Compared with Ir catalyst, which requires an overpotential of ~393 mV to achieve a current density of 10 mA/cm 2 in 0.1 M HClO 4 electrolyte, the evolved Au-Ir catalyst shows a lower overpotential at ~351 mV. X-ray photoelectronic spectrum (XPS) study, in conjunction with electrochemical analysis on the surface-site-normalized activity, reveals that the improved OER performance is due to the presence of Au/Ir interfaces in the catalysts. In addition to Ir, the strategy of accelerating OER via Au/Ir interfaces has been further applied to IrCo, IrNi, and IrCu alloy catalysts. Here, with the broad applicability of the strategy demonstrated, this study opens a new route to design OER catalysts for efficient acidic water splitting.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2016

A domain-specific compiler for a parallel multiresolution adaptive numerical simulation environment

This paper describes the design and implementation of a layered domain-specific compiler to support MADNESS---Multiresolution ADaptive Numerical Environment for Scientific Simulation. MADNESS is a high-level software environment for the solution of integral and differential equations in many dimensions, using adaptive and fast harmonic analysis methods with guaranteed precision. MADNESS uses k-d trees to represent spatial functions and implements operators like addition, multiplication, differentiation, and integration on the numerical representation of functions. The MADNESS runtime system provides global namespace support and a task-based execution model including futures. MADNESS is currently deployed on massively parallel supercomputers and has enabled many science advances. Due to the highly irregular and statically unpredictable structure of the k-d trees representing the spatial functions encountered in MADNESS applications, only purely runtime approaches to optimization have previously been implemented in the MADNESS framework. This paper describes a layered domain-specific compiler developed to address some performance bottlenecks in MADNESS. The newly developed static compile-time optimizations, in conjunction with the MADNESS runtime support, enable significant performance improvement for the MADNESS framework.

Rajbhandari, Samyam↗
DOE OSTI2022

Linkage map construction using limited parental genotypic information

Abstract Genetic linkage maps based on single nucleotide polymorphisms (SNPs) represent an essential tool for a variety of genomic analyses. Today, next-generation sequencing (NGS) enables rapid genotyping of different mapping populations based on thousands of SNPs and the construction of highly saturated linkage maps. Nevertheless, missing data in the genotyping of the parental lines creates a bottleneck that determines the number of SNPs that can be used for the linkage map. As a proof of concept, a highly saturated genetic linkage map was constructed using the imputed genotypic data of a recombinant inbred line (RIL) population and the limited genotypic information of its parental lines. Two ABH genotype files were created from a pseudo-parental genotypic data set that includes all the SNPs present in the RIL population. In the first ABH file pseudo-parental 1 was considered parental A, while in the second pseudo-parental 1 was considered parental B. These two duplicate ABH genotype files were merged by chromosome and subjected to linkage map analysis. Since the ABH data were duplicated, two mirrored linkage groups were generated per chromosome. The correct linkage map was identified and selected based on the partial genotypic data of the parental lines. This strategy was effective for constructing a highly saturated linkage map of 33,421 SNPs based on the genotyping of 205 RILs and a limited number of 100 SNPs present in the parental lines. This strategy enables the use of all the NGS SNP data obtained from a low-coverage sequencing experiment in the mapping population.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1998

A Geometry Based Infra-Structure for Computational Analysis and Design

The computational steps traditionally taken for most engineering analysis suites (computational fluid dynamics (CFD), structural analysis, heat transfer and etc.) are: (1) Surface Generation -- usually by employing a Computer Assisted Design (CAD) system; (2) Grid Generation -- preparing the volume for the simulation; (3) Flow Solver -- producing the results at the specified operational point; (4) Post-processing Visualization -- interactively attempting to understand the results. For structural analysis, integrated systems can be obtained from a number of commercial vendors. These vendors couple directly to a number of CAD systems and are executed from within the CAD Graphical User Interface (GUI). It should be noted that the structural analysis problem is more tractable than CFD; there are fewer mesh topologies used and the grids are not as fine (this problem space does not have the length scaling issues of fluids). For CFD, these steps have worked well in the past for simple steady-state simulations at the expense of much user interaction. The data was transmitted between phases via files. In most cases, the output from a CAD system could go to Initial Graphics Exchange Specification (IGES) or Standard Exchange Program (STEP) files. The output from Grid Generators and Solvers do not really have standards though there are a couple of file formats that can be used for a subset of the gridding (i.e. PLOT3D data formats). The user would have to patch up the data or translate from one format to another to move to the next step. Sometimes this could take days. Specifically the problems with this procedure are:(1) File based -- Information flows from one step to the next via data files with formats specified for that procedure. File standards, when they exist, are wholly inadequate. For example, geometry from CAD systems (transmitted via IGES files) is defined as disjoint surfaces and curves (as well as masses of other information of no interest for the Grid Generator). This is particularly onerous for modern CAD systems based on solid modeling. The part was a proper solid and in the translation to IGES has lost this important characteristic. STEP is another standard for CAD data that exists and supports the concept of a solid. The problem with STEP is that a solid modeling geometry kernel is required to query and manipulate the data within this type of file. (2) 'Good' Geometry. A bottleneck in getting results from a solver is the construction of proper geometry to be fed to the grid generator. With 'good' geometry a grid can be constructed in tens of minutes (even with a complex configuration) using unstructured techniques. Adroit multi-block methods are not far behind. This means that a million node steady-state solution can be computed on the order of hours (using current high performance computers) starting from this 'good' geometry. Unfortunately, the geometry usually transmitted from the CAD system is not 'good' in the grid generator sense. The grid generator needs smooth closed solid geometry. It can take a week (or more) of interaction with the CAD output (sometimes by hand) before the process can begin. One way Communication. (3) One-way Communication -- All information travels on from one phase to the next. This makes procedures like node adaptation difficult when attempting to add or move nodes that sit on bounding surfaces (when the actual surface data has been lost after the grid generation phase). Until this process can be automated, more complex problems such as multi-disciplinary analysis or using the above procedure for design becomes prohibitive. There is also no way to easily deal with this system in a modular manner. One can only replace the grid generator, for example, if the software reads and writes the same files. Instead of the serial approach to analysis as described above, CAPRI takes a geometry centric approach. This makes the actual geometry (not a discretized version) accessible to all phases of the analysis. The connection to the geometry is made through an Application Programming Interface (API) and NOT a file system. This API isolates the top-level applications (grid generators, solvers and visualization components) from the geometry engine. Also this allows the replacement of one geometry kernel with another, without effecting these top-level applications. For example, if UniGraphics is used as the CAD package then Parasolid (UG's own geometry engine) can be used for all geometric queries so that no solid geometry information is lost in a translation. This is much better than STEP because when the data is queried, the same software is executed as used in the CAD system. Therefore, one analyzes the exact part that is in the CAD system. CAPRI uses the same idea as the commercial structural analysis codes but does not specify control. Software components of the CAD system are used, but the analysis suite, not the CAD operator, specifies the control of the software session. This also means that the license issues (may be) minimized and individuals need not have to know how to operate a CAD system in order to run the suite.

Haimes, Robert↗
DOE OSTI2022

Comparative Analysis of Direct and Indirect Cooling of Wide-Bandgap Power Modules and Performance Enhancement of Jet Impingement-Based Direct Substrate Cooling

With the development of high-power and high-torque machines, requirements for high-power density electronics are increasing. Thermal management of such systems requires high heat extraction. Conventional air cooling based heat sinks and cold plate based liquid cooling have their own benefits for various applications but has limitations for high power density applications. The current study explores a jet impingement based direct substrate cooling system that was implemented for a SiC based direct bonded Cu substrate for various power losses. Numerical comparison between jet impingement cooling and conventional horizontal/indirect cooling (pin fin heat sink and genetic algorithm-optimized heat sink) showed that the area weighted average of the heat transfer coefficient (HTC) is high for both horizontal cooling designs, and the local HTC is higher for jet impingement. Design iterations were undertaken to resolve the bottleneck of this cooling system. Increasing the number of nozzles helped to cover more area at the direct bonded Cu bottom plate, which drops the chip temperature considerably. With a constant flow rate, increasing the number of nozzles would decrease local jet velocity, which reduces the heat extraction by jet impingement. This issue can be addressed by reducing the diameter of nozzle but doing so results in a high pressure drop where the design constraint is 2 psi. A flared nozzle design is proposed, which has a higher spreading angle of the jet that increases the flow coverage and reduces the pressure drop of the coolant loop.

Barua, Himel↗
DOE OSTI2020

Chemical Robotics Enabled Exploration of Stability in Multicomponent Lead Halide Perovskites via Machine Learning

Metal halide perovskites have attracted immense interest as a promising material for a variety of optoelectronic and sensing applications. However, issues regarding long-term stability have emerged as the key bottleneck for commercialization. Here, we develop an automated experimental workflow based on combinatorial synthesis and rapid throughput characterization to explore long-term stability of these materials in ambient conditions. We apply it to four model perovskite systems: MA x FA y Cs 1–x–y PbBr 3 , MA x FA y Cs 1–x–y PbI 3 , Cs x FA y MA 1–x–y Pb(Br x+y I 1–x–y ) 3 , and Cs x MA y FA 1–x–y Pb(I x+y Br 1–x–y ) 3 . Non-negative matrix factorization and Gaussian process regression are used to interpolate the photoluminescent behavior of the phase diagram. This interpolative regression analysis helps to distinguish mixtures that form solid solutions from those that segregate into multiple materials, pointing out the most stable regions of the phase diagram. Here, we find stability dependence on composition to be nonuniform within the composition space, suggesting the presence of potential preferential compositional regions. This proposed workflow is universal and can be applied to other solution-processable materials.

36 MATERIALS SCIENCE↗
NASA NTRS1988

Implementation and analysis of a Navier-Stokes algorithm on parallel computers

The results of the implementation of a Navier-Stokes algorithm on three parallel/vector computers are presented. The object of this research is to determine how well, or poorly, a single numerical algorithm would map onto three different architectures. The algorithm is a compact difference scheme for the solution of the incompressible, two-dimensional, time-dependent Navier-Stokes equations. The computers were chosen so as to encompass a variety of architectures. They are the following: the MPP, an SIMD machine with 16K bit serial processors; Flex/32, an MIMD machine with 20 processors; and Cray/2. The implementation of the algorithm is discussed in relation to these architectures and measures of the performance on each machine are given. The basic comparison is among SIMD instruction parallelism on the MPP, MIMD process parallelism on the Flex/32, and vectorization of a serial code on the Cray/2. Simple performance models are used to describe the performance. These models highlight the bottlenecks and limiting factors for this algorithm on these architectures. Finally, conclusions are presented.

Fatoohi, Raad A.↗
DOE OSTI2024

Anode Upcycling via Tailored Solvent Treatment

To achieve a truly closed-loop direct recycling process for lithium-ion batteries, all component materials must be recovered. To date, direct recycling method development has primarily focused on the high-value transition-metal cathode materials, while the inherently lower-value graphite has been challenging to recover in a cost-effective manner. However, end-of-life graphite contains a unique engineered value due to the presence of the solid electrolyte interphase (SEI). Growth of the SEI during the cell's active lifetime stabilizes the electronically reactive graphite surface through an irreversible consumption of Li, and thus necessitates both excess lithiation of the cathode and a costly and time-intensive formation procedure during manufacturing. An optimized pre-formed SEI that capitalizes on existing SEI components from end-of-life batteries has the potential to significantly reduce cathode lithiation requirements and eliminate the critical bottleneck of formation cycling during cell remanufacturing. Further, retaining Li at the anode obviates the need for a separate Li leaching and recovery step, improving the overall efficiency of the direct recycling line. In this work, we present a novel approach to "upcycling" spent graphite through use of tailored chemical treatment to remove adverse (i.e., highly resistive and/or poorly passivating) SEI species while retaining beneficially passivating components. We have explored a rational set of solvents spanning a range of polarity, proticity, and molecular size to evaluate structure-property-performance relationships between applied solvent(s), removed and remaining SEI species, and electrochemical response of the resulting graphite product. Further, we have developed and optimized a robust and holistic analysis procedure that couples symmetric-cell electrochemical testing, multi-modal materials characterization, and advanced electrochemical modeling. These analysis results inform a set of correlative metrics for graphite performance relative to both solvent properties and upcycled SEI composition. We demonstrate effective tunability in the residual SEI composition by varying solvent identity and concentration, and report on several promising solvent systems that achieve comparable or performance to pristine graphite.

anode recycling↗
DOE OSTI2022

Weighted relaxation for multigrid reduction in time

Current trends in computer architectures now mean that faster computation speed must come primarily from increased concurrency, not faster clock speeds, which are stagnating. Thus, this situation creates bottlenecks for serial algorithms, including the well-known bottleneck for sequential time-integration, where each individual time-value (i.e., time-step) is computed sequentially. One approach to alleviate this and achieve parallelism in time is with multigrid. Here, in this work, we consider multigrid-reduction-in-time (MGRIT), a multilevel method applied to the time dimension that computes multiple time-steps in parallel. Like all multigrid methods, MGRIT relies on the complementary relationship between relaxation on a fine-grid and a correction from the coarse grid to solve the problem. All current MGRIT implementations are based on unweighted-Jacobi relaxation; here we introduce the concept of weighted relaxation to MGRIT. We derive new convergence bounds for weighted relaxation, and use this analysis to guide the selection of relaxation weights. Numerical results then demonstrate that by choosing appropriate non-unitary relaxation weights, one can achieve faster convergence rates and lower iteration counts for MGRIT when compared with unweighted relaxation. In most cases, weighted relaxation yields a 10%–20% saving in iterations, which is significant when using large high-performance computers. For A-stable integration schemes, results also illustrate that under-relaxation can restore convergence in some cases where unweighted relaxation is not convergent.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

A new time dependent density functional algorithm for large systems and plasmons in metal clusters

A new algorithm to solve the Time Dependent Density Functional Theory (TDDFT) equations in the space of the density fitting auxiliary basis set has been developed and implemented. The method extracts the spectrum from the imaginary part of the polarizability at any given photon energy, avoiding the bottleneck of Davidson diagonalization. The original idea which made the present scheme very efficient consists in the simplification of the double sum over occupied-virtual pairs in the definition of the dielectric susceptibility, allowing an easy calculation of such matrix as a linear combination of constant matrices with photon energy dependent coefficients. The method has been applied to very different systems in nature and size (from H{sub 2} to [Au{sub 147}]{sup −}). In all cases, the maximum deviations found for the excitation energies with respect to the Amsterdam density functional code are below 0.2 eV. The new algorithm has the merit not only to calculate the spectrum at whichever photon energy but also to allow a deep analysis of the results, in terms of transition contribution maps, Jacob plasmon scaling factor, and induced density analysis, which have been all implemented.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Three practical workflow schedulers for easy maximum parallelism

Runtime scheduling and workflow systems are an increasingly popular algorithmic component in HPC because they allow full system utilization with relaxed synchronization requirements. There are so many special-purpose tools for task scheduling, one might wonder why more are needed. Use cases seen on the Summit supercomputer needed better integration with MPI and greater flexibility in job launch configurations. Preparation, execution, and analysis of computational chemistry simulations at the scale of tens of thousands of processors revealed three distinct workflow patterns. A separate job scheduler was implemented for each one using extremely simple and robust designs: file-based, task-list based, and bulk-synchronous. Comparing to existing methods shows unique benefits of this work, including simplicity of design, suitability for HPC centers, short startup time, and well-understood per-task overhead. All three new tools have been shown to scale to full utilization of Summit, and have been made publicly available with tests and documentation. This work presents a complete characterization of the minimum effective task granularity for efficient scheduler usage scenarios. Here, these schedulers have the same bottlenecks, and hence similar task granularities as those reported for existing tools following comparable paradigms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

An expanded evaluation of protein function prediction methods shows an improvement in accuracy

Background: A major bottleneck in our understanding of the molecular underpinnings of life is the assignment of function to proteins. While molecular experiments provide the most reliable annotation of proteins, their relatively low throughput and restricted purview have led to an increasing role for computational function prediction. However, assessing methods for protein function prediction and tracking progress in the field remain challenging. Results: We conducted the second critical assessment of functional annotation (CAFA), a timed challenge to assess computational methods that automatically assign protein function. We evaluated 126 methods from 56 research groups for their ability to predict biological functions using Gene Ontology and gene-disease associations using Human Phenotype Ontology on a set of 3681 proteins from 18 species. CAFA2 featured expanded analysis compared with CAFA1, with regards to data set size, variety, and assessment metrics. To review progress in the field, the analysis compared the best methods from CAFA1 to those of CAFA2. Conclusions: The top-performing methods in CAFA2 outperformed those from CAFA1. This increased accuracy can be attributed to a combination of the growing number of experimental annotations and improved methods for function prediction. The assessment also revealed that the definition of top-performing algorithms is ontology specific, that different performance metrics can be used to probe the nature of accurate predictions, and the relative diversity of predictions in the biological process and human phenotype ontologies. While there was methodological improvement between CAFA1 and CAFA2, the interpretation of results and usefulness of individual methods remain context-dependent.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Accelerating engineered microbe optimization through machine learning and multi-omics datasets

This project demonstrated the use of a combination of multi-omics data with deep learning and a high-throughput Design- Build-Test-Learn (DBTL) cycle to improve the production of malonic acid, a versatile product with a large market. The project leveraged the unique capabilities of both Lygos and the Agile BioFoundry (ABF): Lygos provided its expertise efficiently designing, building, and cultivating P. kudriavzevii strains; LBNL, PNNL, and NTESS provided multi-omics analysis in the Test phase, LBNL provided machine learning techniques in the Learn phase to analyze the -omics datasets and make recommendations so as to increase malonic acid production in the next DBTL cycle. This project is the first to use large amounts of multi-omics time-series data to feed deep learning models, creating around 80,000 data points in a single DBTL cycle. This project has 1) demonstrated the utility of combining deep learning and multi-omics data sets by improving the production of malonic acid two fold, 2) created a large time-series datasets to be released publicly for external development of new machine learning algorithms, and 3) shown that supply chain problems, strain building bottlenecks, and adaptation times for new ML approaches are key obstacles for fast DBTL cycle times.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

An Open-Source Framework for the Computational Analysis and Design of Autothermal Chemical Processes

The main project object was to develop software tools for simulating non-equilibrium autothermal processes, improving the prospects for identifying and designing such systems. The project demonstrates the use of these tools to simulate autothermal pyrolysis, a process recently developed at the pilot-plant scale at Iowa State University. In such process, instead of externally heating a reactor to pro-vide the enthalpy of pyrolysis, sufficient oxygen in the form of air is introduced into the reactor to support partial oxidation of reactants and products with the exothermic energy released supporting endothermic pyrolysis reactions. A fluidized bed is used to assure good mixing of biomass and oxidant and provide an isothermal reaction environment. The amount of oxygen required depends upon the kind of biomass being pyrolyzed and parasitic heat losses from the reactor. For example, for woody biomass pyrolyzed under conditions that simulate adiabatic operation, equivalence ratios can be as low as 0.06, compared to 0.20 or higher for autothermal gasifiers. By removing the heat transfer bottleneck of conventional pyrolysis, operation in autothermal mode allowed a significant increase in reactor throughput process, approaching five times the throughput of the conventionally operated pyrolyzer. Different simulation strategies were considered and developed: a zero-dimensional chemistry model was used to verify the applicability of kinetic schemes to predict biomass fast pyrolysis in autothermal conditions. Conventional chemical reactor models such as the plug flow reactor and the partially stirred reactor were used to investigate the role of mixing in the fluidized bed pyrolyzer and to establish the impact of mixing time on the gas-phase reactions. A comprehensive multiphase computational fluid dynamics (mCFD) framework, including polydisperse granular phase modeling and detailed chemical kinetics was formulated and used to model the experimental setup for autothermal biomass fast pyrolysis at ISU. Multiphase CFD was also used to investigate the role of biomass feed positioning on the mixing of biomass in the pyrolizer. Finally, a reduced order model (ROM), suitable to be implemented in process simulators was obtained. Both the mCFD and the ROM were validated against experiments.

09 BIOMASS FUELS↗
NASA NTRS2000

On Fast Post-Processing of Global Positioning System Simulator Truth Data and Receiver Measurements and Solutions Data

Post-processing of data, related to a GPS receiver test in a GPS simulator and test facility, is an important step towards qualifying a receiver for space flight. Although the GPS simulator provides all the parameters needed to analyze a simulation, as well as excellent analysis tools on the simulator workstation, post-processing is not a GPS simulator or receiver function alone, and it must be planned as a separate pre-flight test program requirement. A GPS simulator is a critical resource, and it is desirable to move off the pertinent test data from the simulator as soon as a test is completed. The receiver and simulator databases are used to extract the test data files for postprocessing. These files are then usually moved from the simulator and receiver systems to a personal computer (PC) platform, where post-processing is done typically using PC-based commercial software languages and tools. Because of commercial software systems generality their functions are notoriously slow and more than often are the bottleneck even for short duration simulator-based tests. There is a need to do post-processing faster and within an hour after test completion, including all required operations on the simulator and receiver to prepare and move off the post-processing files. This is especially significant in order to use the previous test feedback for the next simulation setup or to run near back-to-back simulation scenarios. Solving the post-processing timing problem is critical for a pre-flight test program success. Towards this goal an approach was developed that allows to speed-up post-processing by an order of a magnitude. It is based on improving the post-processing bottleneck function algorithm using a priory information that is specific to a GPS simulation application and using only the necessary volume of truth data. The presented postprocessing scheme was used in support of a few successful space flight missions carrying GPS receivers.

Kizhner, Semion↗
NASA NTRS1996

Reduction of Weather-Related Terminal Area Delays in the Free-Flight Era

While much of the emphasis of the free-flight movement has been concentrated on reducing en-route delays, airport capacity is a major bottleneck in the current airspace system, particularly during bad weather. According to the Air Transport Association (ATA) Air Carrier Delay Reports, ground delays (gate-hold, taxi-in, and taxi-out) comprise 75 percent of total delays. It is likely that the projected steady growth in traffic will only exacerbate these losses. Preliminary analyses show that implementation of the terminal area technologies and procedures under development in NASA s Terminal Area Productivity program can potentially save the airlines at least $350M annually in weather-related delays by the year 2005 at Boston Logan and Detroit airports alone. This paper briefly describes the Terminal Area Productivity program, outlines the costhenefit analyses that are being conducted in support of the program, and presents some preliminary analysis results.

Johnson, Sally C.↗
DOE OSTI2021

Hot carrier relaxation and inhibited thermalization in superlattice heterostructures: The potential for phonon management

One of the main loss mechanisms in photovoltaic solar cells is the thermalization of photogenerated hot carriers via phonon-mediated relaxation. By inhibiting these relaxation mechanisms and reducing thermalization losses, it may be possible to improve the power conversion efficiency of solar cells beyond the single gap limit. Here, type-II InAs/AlAsSb multi-quantum well (MQW) structures are investigated to study the impact of the phononic properties of the AlAsSb barrier material in hot carrier thermalization. Experimental and theoretical results show that by increasing the barrier thickness (increasing the relative contribution of AlAsSb content in the superlattices), the relaxation of hot carriers is reduced as observed in power-dependent photoluminescence and thermalization analysis. Furthermore, this is attributed to an increase in the phononic bandgap of the MQW with increasing AlAsSb composition reducing the efficiency of the dominant Klemens mechanism as the phononic properties shift toward a more AlSb-like behavior.

14 SOLAR ENERGY↗
DOE OSTI2016

Asynchronous Runtimes in Action: An Introspective Framework for a Next Gen Runtime

One of the most critical challenges that new high performance systems face is the lack of system software support for these large scale systems. Investment on system stack components is essential in the development, debugging and optimization of the new emerging programming models. These emerging models have the promise to better utilize the vast hardware resources available in current and future systems. To aid in the development of applications and new system stacks, runtimes, as instances of their respective execution models, need to produce facilities to introspect their inner workings and allow an indepth attribution of performance bottlenecks and computational patterns. In other words, the runtime systems need to reduce their opacity to observers so that users of a novel program execution model can adapt their designs to fit the intended model usage, regardless of the layer that they are working on. This design/development loop (akin to co-design) enables synergistic opportunities across the entire computational stack. This paper presents the design and implementation of a simple “gray” box performance attribution harness running inside a fine grain runtime system: the Open Community Runtime (OCR). We showcase what such a framework can indicate regarding the runtime behavior while running at scale. To this end, we have designed a set of synthetic scenarios aimed to test the runtime at their best and worst cases. We present an analysis of the most important runtime features, properties and idiosyncrasies that will affect the development of new runtime features, algorithmic selection, and application development.

high performance distributed runtimes↗