Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Programming model”

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 937 records · Page 52

OpenACC acceleration for the P N – P N - 2 algorithm in Nek5000

Due to its high performance and throughput capabilities, GPU-accelerated computing is becoming a popular technology in scientific computing, in particular using programming models such as CUDA and OpenACC. The main advantage with OpenACC is that it enables to simply port codes in their ‘‘original’’ form to GPU systems through compiler directives, thus allowing an incremental approach. An OpenACC implementation is applied to the CFD code Nek5000 for simulation of incompressible flows, based on the spectral-element method. The work follows up previous implementations and focuses now on the P N - P N-2 method for the spatial discretization of the Navier–Stokes equations. Performance results of the ported code show a speed-up of up to 3.1 on multi-GPU for a polynomial order N > 11.

97 MATHEMATICS AND COMPUTING↗

Simit: A Language for Physical Simulation

With existing programming tools, writing high-performance simulation code is labor intensive and requires sacrificing readability and portability. Here, the alternative is to prototype simulations in a high-level language like Matlab, thereby sacrificing performance. The Matlab programming model naturally describes the behavior of an entire physical system using the language of linear algebra. However, simulations also manipulate individual geometric elements, which are best represented using linked data structures like meshes. Translating between the linked data structures and linear algebra comes at significant cost, both to the programmer and to the machine. High-performance implementations avoid the cost by rephrasing the computation in terms of linked or index data structures, leaving the code complicated and monolithic, often increasing its size by an order of magnitude. In this article, we present Simit, a new language for physical simulations that lets the programmer view the system both as a linked data structure in the form of a hypergraph and as a set of global vectors, matrices, and tensors depending on what is convenient at any given time. Simit provides a novel assembly construct that makes it conceptually easy and computationally efficient to move between the two abstractions. Using the information provided by the assembly construct, the compiler generates efficient in-place computation on the graph. We demonstrate that Simit is easy to use: a Simit program is typically shorter than a Matlab program; that it is high performance: a Simit program running sequentially on a CPU performs comparably to hand-optimized simulations; and that it is portable: Simit programs can be compiled for GPUs with no change to the program, delivering 4 to 20× speedups over our optimized CPU code.

97 MATHEMATICS AND COMPUTING↗

Cabana: A Performance Portable Library for Particle-Based Simulations

Particle-based simulations are ubiquitous throughout many fields of computational science and engineering, spanning the atomistic level with molecular dynamics (MD), to mesoscale particle-in-cell (PIC) simulations for solid mechanics, device-scale modeling with PIC methods for plasma physics, and massive N-body cosmology simulations of galaxy structures, with many other methods in between (Hockney & Eastwood, 1989). While these methods use particles to represent significantly different entities with completely different physical models, many low-level details are shared including performant algorithms for short- and/or long-range particle interactions, multi-node particle communication patterns, and other data management tasks such as particle sorting and neighbor list construction. Cabana is a performance portable library for particle-based simulations, developed as part of the Co-Design Center for Particle Applications (CoPA) within the Exascale Computing Project (ECP) (Alexander et al., 2020). The CoPA project and its full development scope, including ECP partner applications, algorithm development, and similar software libraries for quantum MD, is described in (Mniszewski et al., 2021). Cabana uses the Kokkos library for on-node parallelism (Edwards et al., 2014; Trott et al., 2022), enabling simulation on multi-core CPU and GPU architectures, and MPI for GPU-aware, multi-node communication. Cabana provides particle simulation capabilities on almost all current Kokkos backends, including serial execution, OpenMP (including OpenMP-Target for GPUs), CUDA (NVIDIA GPUs), HIP (AMD GPUs), and SYCL (Intel GPUs), providing a clear path for the coming generation of accelerator-based exascale hardware. Cabana builds on Kokkos by providing new particle data structures and particle algorithms resulting in a similar execution policy-based, node-level programming model that is intended to be used in addition to the core Kokkos library within an application. Cabana is designed as an application and physics agnostic, but particle-specific toolkit which can either be used to generate a new application, or to be used as needed in existing applications at various levels of invasiveness including through interfaces that wrap user memory in existing data structures.

97 MATHEMATICS AND COMPUTING↗

Improving the Safety Culture by Strengthening Leadership Skills - 16563

As a primary contractor for the U.S. Department of Energy (DOE), CH2M HILL Plateau Remediation Company (CH2M) is managing a team of approximately 1,400 employees at the Hanford Site in Southeast Washington. The work on this 10-year, $5.7 billion contract ranges from design and construction to decommissioning and cleanup. In 2013, CH2M began the second half of its contract and expanded its leadership development program to focus on succession planning and strengthening the leadership skills of its employees. The goal is to create and sustain qualified leaders to lead the Hanford cleanup mission today and into the future. CH2M launched the Leadership Impact Initiative, a training program modeled from best practices used at other CH2M projects, including the Idaho Cleanup Project. Based on the results of a CH2M HILL Plateau Remediation Company site-wide safety culture survey, the company developed an improvement strategy focused on leadership development with emphasis on the front-line leadership. The strategy stresses core leadership principles and skill development to help managers more effectively engage the workforce. The two-day workshop gives employees the opportunity to build leadership on the job in an audience of their peers. CH2M expanded the workshops to include employees from all levels of the organization, including bargaining unit employees. The workshops are reinforced at quarterly all-manager meetings, communications among team members and regular sharing of lessons learned and experiences in the field. The Leadership Impact Initiative advances CH2M's vision of quality leadership and project delivery anchored by health and safety performance excellence. Since 2013, CH2M hosted more than 25 Leadership Impact Initiative workshops, reaching approximately 500 participants. Meanwhile, the company achieved the best safety record since the start of its contract, and results from its 2015 safety culture survey show improvements in the safety climate. A DOE assessment of the CH2M worker health and safety culture for the Voluntary Protection Program (VPP) found the workshop was a contributing factor to the contractor achieving VPP Star and VPP Star of Excellence Status. This paper summarizes the strategy behind the leadership development program, the outcomes and its relevance for other companies, particularly those on DOE Environmental Management decommissioning projects. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Capacitated location of collection sites in an urban waste management system

Urban waste management is becoming an increasingly complex task, absorbing a huge amount of resources, and having a major environmental impact. The design of a waste management system consists in various activities, and one of these is related to the location of waste collection sites. In this paper, we propose an integer programming model that helps decision makers in choosing the sites where to locate the unsorted waste collection bins in a residential town, as well as the capacities of the bins to be located at each collection site. This model helps in assessing tactical decisions through constraints that force each collection area to be capacitated enough to fit the expected waste to be directed to that area, while taking into account Quality of Service constraints from the citizens' point of view. Moreover, we propose an effective constructive heuristic approach whose aim is to provide a good solution quality in an extremely reduced computational time. Computational results on data related to the city of Nardo, in the south of Italy, show that both exact and heuristic approaches provide consistently better solutions than that currently implemented, resulting in a lower number of activated collection sites, and a lower number of bins to be used.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

HEP Science Network Requirements--Final Report

The Energy Sciences Network (ESnet) is the primary provider of network connectivity for the US Department of Energy Office of Science, the single largest supporter of basic research in the physical sciences in the United States. In support of the Office of Science programs, ESnet regularly updates and refreshes its understanding of the networking requirements of the instruments, facilities, scientists, and science programs that it serves. This focus has helped ESnet to be a highly successful enabler of scientific discovery for over 20 years. In August 2009 ESnet and the Office of High Energy Physics (HEP), of the DOE Office of Science, organized a workshop to characterize the networking requirements of the programs funded by HEP. The International HEP community has been a leader in data intensive science from the beginning. HEP data sets have historically been the largest of all scientific data sets, and the communty of interest the most distributed. The HEP community was also the first to embrace Grid technologies. The requirements identified at the workshop are summarized below, and described in more detail in the case studies and the Findings section: (1) There will be more LHC Tier-3 sites than orginally thought, and likely more Tier-2 to Tier-2 traffic than was envisioned. It it not yet known what the impact of this will be on ESnet, but we will need to keep an eye on this traffic. (2) The LHC Tier-1 sites (BNL and FNAL) predict the need for 40-50 Gbps of data movement capacity in 2-5 years, and 100-200 Gbps in 5-10 years for HEP program related traffic. Other key HEP sites include LHC Tier-2 and Tier-3 sites, many of which are located at universities. To support the LHC, ESnet must continue its collaborations with university and international networks. (3) While in all cases the deployed 'raw' network bandwidth must exceed the user requirements in order to meet the data transfer and reliability requirements, network engineering for trans-Atlantic connectivity is more complex than network engineering for intra-US connectivity. This is because transoceanic circuits have lower reliability and longer repair times when compared with land-based circuits. Therefore, trans-Atlantic connectivity requires greater deployed bandwidth and diversity to ensure reliability and service continuity of the user-level required data transfer rates. (4) Trans-Atlantic traffic load and patterns must be monitored, and projections adjusted if necessary. There is currently a shutdown planned for the LHC in 2012 that may affect projections of trans-Atlantic bandwidth requirements. (5) There is a significant need for network tuning and troubleshooting during the establishment of new LHC Tier-2 and Tier-3 facilities. ESnet will work with the HEP community to help new sites effectively use the network. (6) SLAC is building the CCD camera for the LSST. This project will require significant bandwidth (up to 30Gbps) to NCSA over the next few years. (7) The accelerator modeling program at SLAC could require the movement of 1PB simulation data sets from the Leadership Computing Facilities at Argonne and Oak Ridge to SLAC. The data sets would need to be moved overnight, and moving 1PB in eight hours requires more than 300Gbps of throughput. This requirement is dependent on the deployment of analysis capabilities at SLAC, and is about five years away. (8) It is difficult to achieve high data transfer throughput to sites in China. Projects that need to transfer data in or out of China are encouraged to deploy test and measurement infrastructure (e.g. perfSONAR) and allow time for performance tuning.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Lake Whitney Comprehensive Water Quality Assessment, Phase 1B- Physical and Biological Assessment (USDOE)

Baylor University Center for Reservoir and Aquatic Systems Research (CRASR) has conducted a phased, comprehensive evaluation of Lake Whitney to determine its suitability for use as a regional water supply reservoir. The area along the Interstate 35 corridor between Dallas / Fort Worth Metroplex and the Waco / Temple Centroplex represents one of the fastest growth areas in the State of Texas and reliable water supplies are critical to sustainable growth. Lake Whitney is situated midway between these two metropolitan areas. Currently, the City of Whitney as well as all of Bosque and Hill counties obtain their potable water from the Trinity Sands aquifer. Additionally, parts of the adjoining McLennan and Burleson counties utilize the Trinity sands aquifer system as a supplement to their surface water supplies. Population growth coupled with increasing demands on this aquifer system in both the Metroplex and Centroplex have resulted in a rapid depletion of groundwater in these rural areas. The Lake Whitney reservoir represents both a potentially local and regional solution for an area experiencing high levels of growth. Because of the large scope of this project as well as the local, regional and national implications, we have designed a multifaceted approach that will lead to the solution of numerous issues related to the feasibility of using Lake Whitney as a water resource to the region. Phase IA (USEPA, QAPP Study Elements 1-4) of this research focused on the physical limnology of the reservoir (bathymetry and fine scale salinity determination) and develops hydrodynamic watershed and reservoir models to evaluate how salinity would be expected to change with varying hydrologic and climatic factors. To this end, we implemented a basic water quality modeling program in collaboration with the Texas Parks and Wildlife Department and the Texas Commission on Environmental Quality to add to the developing long-term database on Lake Whitney. Finally, we conducted an initial assessment of knowledge of watershed and water quality related issues by local residents and stakeholders of Lake Whitney and design an intervention educational program to address any deficiencies discovered. Phase IA was funded primarily from EPA Cooperative Agreement X7-9769 8901-0. Phase IC (USEPA, QAPP Study Element 5) of this research focused on the ambient toxicity of the reservoir with respect to periodic blooms of golden algae. Phase IC was funded primarily from Cooperative Agreement EM-96638001. Phase 1B (USDOE, Study Elements 6-11) complemented work being done via EPA funding on study elements 1-5 and added five new study elements: 6) Salinity Transport in the Brazos Watershed to Lake Whitney; 7) Bacterial Assessment; 8) Organic Contaminant Analysis on Lake Whitney; 9) Plankton Photosynthesis; 10) Lake Whitney Resident Knowledge Assessment; and 11) Engineering Scoping Perspective: Recommendations for Use.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Distributed biomass supply chain cost optimization to evaluate multiple feedstocks for a biorefinery

Conventional practices of siting all biomass preprocessing operations at the biorefinery is widely believed to be the most cost-effective solution for feedstock supply because of economies of scale. Yet, biomass preprocessing operations could be decentralized by moving the preprocessing operations to distributed biomass preprocessing centers, also known as “depots” located near biomass sources. We present a comparative case study with multiple biomass resources to analyze biorefinery feedstock supply logistics designs having distributed depots and a primary depot co-located with the biorefinery. A mixed-integer linear programming model was developed to simultaneously optimize feedstock sourcing decisions, and optimal preprocessing depot locations and size, utilizing biomass resources from agricultural residue, energy and municipal solid waste to meet carbohydrate specifications and feedstock demand for a biochemical conversion process. Findings from a case study in the US showed that a biorefinery could increase its feedstock supply draw area and supply volume by 57.3%, 177.4% respectively without increasing the feedstock delivered cost by adopting distributed depot-in the feedstock supply chain design. A distributed-depot-based supply chain can be more economical by selecting optimal mix of biomass resource, optimal siting and depot scales during feedstock supply chain design. The results from this study indicate that a biorefinery can utilize dynamic blending to meet the feedstock quality specifications as well as larger supply radius in the distributed depot-based supply chain design to access more available biomass to handle potential feedstock supply uncertainty.

09 BIOMASS FUELS↗

Optimization of a Mixed Fleet of Aerial Drones for Medical Supplies: A Case Study of Blood Delivery Logistics

Aerial drones have emerged as an innovative solution for faster transportation of time-sensitive items (e.g., emergency medical supplies), potentially reducing the transmission of contagious diseases and enhancing healthcare availability through contactless autonomous delivery. We study fleet sizing and efficient scheduling of a mixed fleet of drones for delivering time-sensitive medical items having distinct release and due times to minimize the required fleet size and fleet composition, the required number of additional batteries, and the total energy consumption. We continuously track the remaining battery energy of drones to determine the optimal timing for battery replacement, rather than replacing the battery at each node. Using actual drone flight test data, we employed a machine learning (ML) method to estimate the energy consumption of different drone types during flight segments for different operating parameters. We present a novel mixed-integer programming model to efficiently formulate the problem that integrates the estimated energy consumption functions from ML. We propose a new greedy heuristic (GH) algorithm and a customized genetic algorithm (GA) for solving large-scale instances of this problem faster. Results demonstrate that the GH algorithm is substantially faster than the accelerated CPLEX and the GA, while sacrificing the solution quality by a small amount. Results based on an actual blood sample delivery case study from Pendleton, Oregon, United States, show that using a mixed fleet of drones reduces the total cost and total energy consumption up to 18.18% and 28.7%, respectively, compared to using a homogeneous fleet.

29 - ENERGY PLANNING, POLICY AND ECONOMY↗

Aerial drone fleet deployment optimization with endogenous battery replacements for direct delivery of time-sensitive products

Aerial drones offer a distinct potential to reduce the delivery time and energy consumption for the delivery of time-sensitive and small products. However, there is still a need in the relevant industry to understand the performance of drone-based delivery under different business needs and drone operating conditions. We studied a drone deployment optimization problem for direct delivery of time-sensitive products with release dates to customers maintaining a specified time window. This paper presents a new mixed-integer programming model, new valid inequalities, a new greedy heuristic algorithm, and a Genetic algorithm to help business owners optimally schedule and route their drone fleet minimizing the required fleet size, the required number of additional batteries, and total energy consumption. A realistic feature of the optimization method is that instead of replacing the drone battery after each return to the depot, it keeps track of the remaining energy in the drone battery and decides on battery replacements accounting for the drone routing and the user-specified minimum required battery energy. Numerical results based on real data from drone flight tests and prepared food delivery industry provide insights into the effect of different practical drone operating parameters on the required fleet size, the required number of battery replacements, and energy consumption. Here, results demonstrate that the proposed heuristic algorithm substantially outperforms the accelerated CPLEX in runtime while sacrificing the solution quality by a small amount. Additionally, results show that using a mixed fleet of hexacopter and quadcopter drones reduces the total energy consumption by 48.52% compared to using a homogeneous fleet of only hexacopters.

Drone energy consumption↗

A ROSE-based OpenMP 3.0 Research Compiler Supporting Multiple Runtime Libraries

OpenMP is a popular and evolving programming model for shared-memory platforms. It relies on compilers for optimal performance and to target modern hardware architectures. A variety of extensible and robust research compilers are key to OpenMP's sustainable success in the future. In this paper, we present our efforts to build an OpenMP 3.0 research compiler for C, C++, and Fortran; using the ROSE source-to-source compiler framework. Our goal is to support OpenMP research for ourselves and others. We have extended ROSE's internal representation to handle all of the OpenMP 3.0 constructs and facilitate their manipulation. Since OpenMP research is often complicated by the tight coupling of the compiler translations and the runtime system, we present a set of rules to define a common OpenMP runtime library (XOMP) on top of multiple runtime libraries. These rules additionally define how to build a set of translations targeting XOMP. Our work demonstrates how to reuse OpenMP translations across different runtime libraries. This work simplifies OpenMP research by decoupling the problematic dependence between the compiler translations and the runtime libraries. We present an evaluation of our work by demonstrating an analysis tool for OpenMP correctness. We also show how XOMP can be defined using both GOMP and Omni and present comparative performance results against other OpenMP compilers.

99 GENERAL AND MISCELLANEOUS↗

Data-Parallel Halo Finder Operator in PISTON

PISTON is a portable framework which supports the development of visualization and analysis operators using a platform-independent, data-parallel programming model. Operators such as isosurface, cut-surface and threshold have been implemented in this framework, with the exact same operator code achieving good parallel performance on different architectures. An important analysis operator in cosmology is the halo finder. A halo is a cluster of particles and is considered a common feature of interest found in cosmology data. As the number of cosmological simulations carried out in the recent past has increased, the resultant data of these simulations and the required analysis tasks have increased as well. As a consequence, there is a need to develop scalable and efficient tools to carry out the needed analysis. Therefore, we are currently implementing a halo finder operator using PISTON. Researchers have developed a wide variety of techniques to identify halos in raw particle data. The most basic algorithm is the friend-of-friends (FOF) halo finder, where the particles are clustered based on two parameters: linking length and halo size. In a FOF halo finder, all particles which lie within the linking length are considered as one halo and the halos are filtered based on the halo size parameter. A naive implementation of a FOF halo finder compares each and every particle pair, requiring O(n{sup 2}) operations. Our data-parallel halo finder operator uses a balanced k-d tree to reduce this number of operations in the average case, and implements the algorithm using only the data-parallel primitives in order to achieve portability and performance.

79 ASTRONOMY AND ASTROPHYSICS↗

Integrated assessment of a new Waste-to-Energy facility in Central Greece in the context of regional perspectives

The main aim of this study is the integrated assessment of a proposed Waste-to-Energy facility that could contribute in the Municipal Solid Waste Management system of the Region of Central Greece. In the context of this paper alternative transfer schemes for supplying the candidate facility were assessed considering local conditions and economical criteria. A mixed-integer linear programming model was applied for the determination of optimum locations of Transfer Stations for an efficient supplying chain between the waste producers and the Waste-to-Energy facility. Moreover different Regional Waste Management Scenarios were assessed against multiple criteria, via the Multi Criteria Decision Making method ELECTRE III. The chosen criteria were total cost, Biodegradable Municipal Waste diversion from landfill, energy recovery and Greenhouse Gas emissions and the analysis demonstrated that a Waste Management Scenario based on a Waste-to-Energy plant with an adjacent landfill for disposal of the residues would be the best performing option for the Region, depending however on the priorities of the decision makers. In addition the study demonstrated that efficient planning is necessary and the case of three sanitary landfills operating in parallel with the WtE plant in the study area should be avoided. Moreover alternative cases of energy recovery of the candidate Waste-to-Energy facility were evaluated against the requirements of the new European Commission Directive on waste in order for the facility to be recognized as recovery operation. The latter issue is of high significance and the decision makers in European Union countries should take it into account from now on, in order to plan and implement facilities that recover energy efficiently. Finally a sensitivity check was performed in order to evaluate the effects of increased recycling rate, on the calorific value of treated Municipal Solid Waste and the gate fee of the candidate plant and found that increased recycling efforts would not diminish the potential for incineration with energy recovery from waste and neither would have adverse impacts on the gate fee of the Waste-to-Energy plant. In general, the study highlighted the need for efficient planning in solid waste management, by taking into account multiple criteria and parameters and utilizing relevant tools and methodologies into this context.

09 BIOMASS FUELS↗

A review on regional convection‐permitting climate modeling: Demonstrations, prospects, and challenges

Abstract Regional climate modeling using convection‐permitting models (CPMs; horizontal grid spacing <4 km) emerges as a promising framework to provide more reliable climate information on regional to local scales compared to traditionally used large‐scale models (LSMs; horizontal grid spacing >10 km). CPMs no longer rely on convection parameterization schemes, which had been identified as a major source of errors and uncertainties in LSMs. Moreover, CPMs allow for a more accurate representation of surface and orography fields. The drawback of CPMs is the high demand on computational resources. For this reason, first CPM climate simulations only appeared a decade ago. In this study, we aim to provide a common basis for CPM climate simulations by giving a holistic review of the topic. The most important components in CPMs such as physical parameterizations and dynamical formulations are discussed critically. An overview of weaknesses and an outlook on required future developments is provided. Most importantly, this review presents the consolidated outcome of studies that addressed the added value of CPM climate simulations compared to LSMs. Improvements are evident mostly for climate statistics related to deep convection, mountainous regions, or extreme events. The climate change signals of CPM simulations suggest an increase in flash floods, changes in hail storm characteristics, and reductions in the snowpack over mountains. In conclusion, CPMs are a very promising tool for future climate research. However, coordinated modeling programs are crucially needed to advance parameterizations of unresolved physics and to assess the full potential of CPMs.

54 ENVIRONMENTAL SCIENCES↗

Adoption of biofuels for marine shipping decarbonization: A long‐term price and scalability assessment

Abstract This study assessed the long‐term annual biofuel production capacity potential and price in the United States and shed light on the prospect of biofuel adoption for marine propulsion. A linear programming model was developed to assist the projections and provide insightful analyses. The projected long‐term (2040) maximum annual capacity of biofuels in the United States is 245 million metric tons (Mt) or 65 billion gallons of heavy fuel oil gallon equivalent (HFOGE) when based on the median feedstock availability. Between 2022 (near‐term) and 2040, the potential biofuel capacity increases by over 40%, attributed to increased feedstock availability. At a price range up to $500/t, biodiesel is the main product, and the annual capacity (12 Mt) is limited to feedstock availability constraints. Biodiesel and corn ethanol are the main biofuels at a price range up to $750/t. At a higher price point (above $750/t), the biofuel types and annual capacities increase substantially (218 Mt per year). Biofuels above this price include gasoline‐, jet‐, and diesel‐range blendstocks, as well as bio‐methanol, bio‐propane, and biogas. This study concludes that the US domestic feedstock availability coupled with advanced conversion technologies can produce substantial amounts of biofuels to achieve a critical mass and be impactful as alternative marine fuels. There is also a need to improve the biofuel price for marine shipping adoption. Policies and economic incentives that provide temporary financial support would help facilitate maritime biofuel adoption. © 2022 Alliance for Sustainable Energy, LLC. Biofuels, Bioproducts and Biorefining published by Society of Industrial Chemistry and John Wiley & Sons Ltd.

09 BIOMASS FUELS↗

Performance analysis of distributed symmetric sparse matrix vector multiplication algorithm for multi-core architectures

In this article, sparse matrix vector multiply (SpMVM) is an important kernel that frequently arises in high performance computing applications. Due to its low arithmetic intensity, several approaches have been proposed in literature to improve its scalability and efficiency in large scale computations. In this paper, our target systems are high end multi-core architectures and we use messaging passing interface + open multiprocessing hybrid programming model for parallelism. We analyze the performance of recently proposed implementation of the distributed symmetric SpMVM, originally developed for large sparse symmetric matrices arising in ab initio nuclear structure calculations. We also study important features of this implementation and compare with previously reported implementations that do not exploit underlying symmetry. Our SpMVM implementations leverage the hybrid paradigm to efficiently overlap expensive communications with computations. Our main comparison criterion is the "CPU core hours" metric, which is the main measure of resource usage on supercomputers. We analyze the effects of topology-aware mapping heuristic using simplified network load model. Furthermore, we have tested the different SpMVM implementations on two large clusters with 3D Torus and Dragonfly topology. Our results show that the distributed SpMVM implementation that exploits matrix symmetry and hides communication yields the best value for the "CPU core hours" metric and significantly reduces data movement overheads.

97 MATHEMATICS AND COMPUTING↗

Load‐balanced and locality‐aware scheduling for data‐intensive workloads at extreme scales

Summary Data‐driven programming models such as many‐task computing (MTC) have been prevalent for running data‐intensive scientific applications. MTC applies over‐decomposition to enable distributed scheduling. To achieve extreme scalability, MTC proposes a fully distributed task scheduling architecture that employs as many schedulers as the compute nodes to make scheduling decisions. Achieving distributed load balancing and best exploiting data locality are two important goals for the best performance of distributed scheduling of data‐intensive applications. Our previous research proposed a data‐aware work‐stealing technique to optimize both load balancing and data locality by using both dedicated and shared task ready queues in each scheduler. Tasks were organized in queues based on the input data size and location. Distributed key‐value store was applied to manage task metadata. We implemented the technique in MATRIX, a distributed MTC task execution framework. In this work, we devise an analytical suboptimal upper bound of the proposed technique, compare MATRIX with other scheduling systems, and explore the scalability of the technique at extreme scales. Results show that the technique is not only scalable but can achieve performance within 15% of the suboptimal solution. Copyright © 2015 John Wiley & Sons, Ltd.

Wang, Ke↗