Search NASA⌕ Search

SEARCH · Search NASA

Results for “Load Balancing”

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 181 records · Page 10

Implementation of the chemistry module MECCA (v2.5) in the modal aerosol version of the Community Atmosphere Model component (v3.6.33) of the Community Earth System Model

A coupled atmospheric chemistry and climate system model was developed using the modal aerosol version of the National Center for Atmospheric Research Community Atmosphere Model (modal-CAM; v3.6.33) and the Max Planck Institute for Chemistry’s Module Efficiently Calculating the Chemistry of the Atmosphere (MECCA; v2.5) to provide enhanced resolution of multiphase processes, particularly those involving inorganic halogens, and associated impacts on atmospheric composition and climate. Three Rosenbrock solvers (Ros-2, Ros-3, RODAS-3) were tested in conjunction with the basic load-balancing options available to modal-CAM (1) to establish an optimal configuration of the implicitly-solved multiphase chemistry module that maximizes both computational speed and repeatability of Ros- 2 and RODAS-3 results versus Ros-3, and (2) to identify potential implementation strategies for future versions of this and similar coupled systems. RODAS-3 was faster than Ros-2 and Ros-3 with good reproduction of Ros-3 results, while Ros-2 was both slower and substantially less reproducible relative to Ros-3 results. Modal-CAM with MECCA chemistry was a factor of 15 slower than modal-CAM using standard chemistry. MECCA chemistry integration times demonstrated a systematic frequency distribution for all three solvers, and revealed that the change in run-time performance was due to a change in the frequency distribution of chemical integration times; the peak frequency was similar for all solvers. This suggests that efficient chemistry-focused load-balancing schemes can be developed that rely on the parameters of this frequency distribution.

Long, M S↗

ESnet/JLab FPGA Accelerated Transport

To increase the science rate for high data rates/volumes, Thomas Jefferson National Accelerator Facility (JLab) has partnered with Energy Sciences Network (ESnet) to define an edge to data center traffic shaping / steering transport capability featuring data event aware network shaping and forwarding. The keystone of this ESnet+JLab FPGA Accelerated Transport (EJFAT) is the joint development of an AI/ML directed dynamic compute work Load Balancer (LB) of UDP streamed data. The LB is a suite consisting of a Field Programmable Gate Array (FPGA) executing the dynamically configurable, low fixed latency LB data plane featuring real-time packet redirection and high throughput, and a control plane running on the FPGA host computer that monitors network and compute farm telemetry in order to make dynamic AI/ML guided decisions for destination compute host redirection/load balancing and destination resource provisioning. The LB provides for three-tier horizontal scaling across LB suites, core compute hosts, and CPUs within a host. The LB effectively provides seamless integration of edge/core computing to support direct experimental data processing for immediate use by JLab science programs and others such as the EIC as well as data centers of the future requiring high throughput and low latency for both hot and cooled data for both running experiment data acquisition systems and data center use cases.

97 MATHEMATICS AND COMPUTING↗

FPDeep: Scalable Acceleration of CNN Training on Deeply-Pipelined FPGA Clusters

In this paper, we propose a framework called FPDeep, which uses a hybrid of model and layer paral- lelism to configure distributed reconfigurable clusters to train DNNs. This approach has numerous benefits. First, the design does not suffer from batch size growth. Second, novel workload and weight partitioning leads to balanced loads of both among nodes. And third, the entire system is fine-grained pipeline. This leads to high parallelism and utilization and also minimizes the time features need to be cached while waiting for back-propagation.

Wang, Tianqi↗

Automatic and Transparent Resource Contention Mitigation for Improving Large-Scale Parallel File System Performance

Proportional to the scale increases in HPC systems, many scientific applications are becoming increasingly data intensive, and parallel I/O has become one of the dominant factors impacting the large-scale HPC application performance. On a typical large-scale HPC system, we have observed that the lack of a global workload coordination coupled with the shared nature of storage systems cause load imbalance and resource contention over the end-to-end I/O paths resulting in severe performance degradation. I/O load imbalance on HPC systems is generally a self-inflicted wound and mostly occurs between the I/O paths and resources consumed by each individual job. In this paper, we introduce TAPP-IO, a dynamic, shared load balancing framework for mitigating resource contention. TAPP-IO extends our previous work and solves two major limitations: First, it transparently intercepts file creation calls during runtime to balance the workload over all available storage targets. The usage of TAPP-IO requires no application source code modifications and is independent from any I/O middleware. The framework can be applied to almost any HPC platform and is suitable for systems that lack a centralized file system resource manager. Second, the framework proposes a new placement strategy to support not only file-per-process I/O, but also single shared file I/O. This opens the door to a new class of scientific applications that can leverage the placement library for improved performance. We demonstrate the effectiveness of our integration on the Titan system at the Oak Ridge National Laboratory. Our experiments with a synthetic benchmark and real-world HPC workload show that, even in a noisy production environment, TAPP-IO can improve large-scale application performance significantly.

Neuwirth, Sarah↗

Performance analysis and optimization for scalable deployment of deep learning models for country-scale settlement mapping on Titan supercomputer

Here, we present a scalable object detection workflow for detecting objects, such as settlements, from remotely sensed (RS) imagery. We have successfully deployed this workflow on Titan supercomputer and utilized it for the task of mapping human settlement at a country scale. The performance of various stages in the workflow was analyzed before making it operational. The workflow implemented various strategies to address issues such as suboptimal resource utilization and long-tail effects due to unbalanced image workload, data loss due to runtime failures, and maximum wall-time constraints imposed by Titan's job scheduling policy. A mean shift clustering–based static load balancing strategy was implemented, which partitions the image load such that each partition contained similar-sized images. Furthermore, a checkpoint-restart strategy was added in the workflow as a fault-tolerance mechanism to prevent the data losses due to unforeseen runtime failures. The performance of the above-mentioned strategies was observed in various scenarios, such as node failure, exceeding wall time, and successful completion. Using this workflow, we have examined an RS data set that has a spatial resolution of 0.31 m and is comprised of 685 675 km 2 of area of the Republic of Zambia in under six hours using 5426 nodes of the Titan supercomputer.

97 MATHEMATICS AND COMPUTING↗

Performance Analysis of Speculative Parallel Adaptive Local Timestepping for Conservation Laws

Stable simulation of conservation laws, such as those used to model fluid dynamics and plasma physics applications, requires the satisfaction of the so-called Courant-Friedrichs-Lewy condition. By allowing regions of the mesh to advance with different timesteps that locally satisfy this stability constraint, significant work reduction can be attained when compared to a time integration scheme using a single timestep size. However, parallelizing this algorithm presents considerable difficulty. Since the stability condition depends on the state of the system, dependencies become dynamic and potentially non-local. In this article, we present an adaptive local timestepping algorithm using an optimistic (Timewarp-based) parallel discrete event simulation. We introduce waiting heuristics to limit misspeculation and a semi-static load balancing scheme to eliminate load imbalance as parts of the mesh require finer or coarser timesteps. Last, we outline an interface for separating the physics of the specific conservation law from the temporal integration allowing for productive adoption of our proposed algorithm. We present a misspeculation study for three conservation laws, demonstrating both the productivity of the local timestepping API, for which 74% of the lines of code are reused across different conservation laws, and the robustness of the waiting heuristics—at most 1.5% of element updates are rolled back. Our performance studies demonstrate up to a 2.8× speedup versus a baseline unoptimized local timestepping approach, a 4x improvement in per-node throughput compared to an MPI parallelization of synchronous timestepping, and scalability up to 3,072 cores on NERSC’s Cori Haswell partition.

97 MATHEMATICS AND COMPUTING↗

Scaling Demand Flexibility: Building on 30 Years of Energy Efficiency Success

With electricity consumption across the United States (US) and Canada anticipated to grow, energy efficiency program administrators have a key role to play in helping to ensure energy affordability and reliability in support of the broader economic systems utilities and grid support. Connected, demand side load balancing solutions, such as load shifting heating, ventilation and air conditioning (HVAC) systems and managed charging for electric vehicles (EVs), can dynamically manage energy, allowing for more volumetric electricity consumption without incurring the expense of upgraded transmission and distribution capabilities. When combined, or aggregated, many small loads can be managed to have meaningful impact on energy demand on the grid. Utilities and their partners have an opportunity to leverage decades of experience and the infrastructure needed to assess, design, implement, and measure programs to scale up the adoption of equipment with built-in load flexibility capabilities. Current efforts among a wide variety of electricity system service providers, utilities, standards agencies, regulators, national labs and private industry stakeholders aim to identify common standards, metrics, and methodologies for valuing grid services offered by demand side equipment. By combining those efforts with decades of proven energy efficiency resources, utilities are poised to effectuate a scaling up of equipment with energy management capabilities installed in homes and businesses across the US and Canada. This paper will provide an overview of how utilities are approaching this era of load growth and new peak demands across the United States and Canada. It will highlight the specific strategies that program administrators are employing to advance market transformation for grid-enabled products and devices that have the greatest potential to reduce energy use and increase load flexibility.

Grant, Peter↗

Did the GPU obfuscate the load imbalance in my MPI simulation?

The current proliferation of GPU-based HPC systems necessitates a method for assessing the performance of simulations on heterogeneous machines. The addition of GPUs to a system adds multiple hierarchical levels of parallelism to the node architecture. In this paper, we demonstrate that the traditional load imbalance metric is insufficient for capturing the load imbalance on GPU-based machines, since it treats the GPU as a monolithic entity and ignores the internal parallelism. We propose a new hierarchical metric that improves the correlation of measured performance and application workload by up to 20.61%. Using our metric for determining application load instead of the traditional metric as the input for the load balancing algorithm reduces the residual load imbalance by up to 4× in our application.

Eberius, David↗

The Role of Energy Storage in Commercial Building

Motivation and Background of Study This project was motivated by the need to understand the full value of energy storage (thermal and electric energy storage) in commercial buildings, the opportunity of benefits for building operations and the potential interactions between a building and a smart grid infrastructure. On-site or local energy storage systems are not new to the commercial building sector; they have been in place in US buildings for decades. Most building-scale storage technologies are based on thermal or electrochemical storage mechanisms. Energy storage technologies are not designed to conserve energy, and losses associated with energy conversion are inevitable. Instead, storage provides flexibility to manage load in a building or to balance load and generation in the power grid. From the building owner's perspective, storage enables load shifting to optimize energy costs while maintaining comfort. From a grid operations perspective, building storage at scale could provide additional flexibility to grid operators in managing the generation variability from intermittent renewable energy resources (wind and solar). To characterize the set of benefits, technical opportunities and challenges, and potential economic values of storage in a commercial building from both the building operation's and the grid operation's view-points is the key point of this project. The research effort was initiated in early 2010 involving Argonne National Laboratory (ANL), the National Renewable Energy Laboratory (NREL), and Pacific Northwest National Laboratory (PNNL) to quantify these opportunities from a commercial buildings perspective. This report summarizes the early discussions, literature reviews, stakeholder engagements, and initial results of analyses related to the overall role of energy storage in commercial buildings. Beyond the summary of roughly eight months of effort by the laboratories, the report attempts to substantiate the importance of active DOE/BTP R&D activities in this space.

14 SOLAR ENERGY↗

Final Report: Migration Mechanisms for Large-scale Parallel Applications

Process migration is the ability to transfer a process from one machine to another. It is a useful facility in distributed computing environments, especially as computing devices become more pervasive and Internet access becomes more ubiquitous. The potential benefits of process migration, among others, are fault resilience by migrating processes off of faulty hosts, data access locality by migrating processes closer to the data, better system response time by migrating processes closer to users, dynamic load balancing by migrating processes to less loaded hosts, and improved service availability and administration by migrating processes before host maintenance so that applications can continue to run with minimal downtime. Although process migration provides substantial potential benefits and many approaches have been considered, achieving transparent process migration functionality has been difficult in practice. To address this problem, our work has designed, implemented, and evaluated new and powerful transparent process checkpoint-restart and migration mechanisms for desktop, server, and parallel applications that operate across heterogeneous cluster and mobile computing environments. A key aspect of this work has been to introduce lightweight operating system virtualization to provide processes with private, virtual namespaces that decouple and isolate processes from dependencies on the host operating system instance. This decoupling enables processes to be transparently checkpointed and migrated without modifying, recompiling, or relinking applications or the operating system. Building on this lightweight operating system virtualization approach, we have developed novel technologies that enable (1) coordinated, consistent checkpoint-restart and migration of multiple processes, (2) fast checkpointing of process and file system state to enable restart of multiple parallel execution environments and time travel, (3) process migration across heterogeneous software environments, (4) network checkpoint-restart and migration of distributed and parallel applications, (5) a utility computing infrastructure for mobile desktop cloud computing based on process checkpoint-restart and migration functionality, (6) a process migration security architecture for protecting applications and infrastructure from denial-of-service attacks, and (7) a checkpoint-restart mobile computing system using portable storage devices.

99 GENERAL AND MISCELLANEOUS↗

Coordinated Optimal Control of Constrained DERs

In this paper, a coordinated optimal control is designed to simultaneously satisfy multiple power system objectives at the distribution network level. With high penetration of renewables, traditional generation and control may not be enough to compensate for intermittent changes in net load. Dynamic coordination of battery storage devices and demand response at aggregated distributed energy resource (DER) sites becomes necessary to maintain power system operation, however, each dynamic power component operates at different time scales. Frequency stabilization can be achieved by designing optimal controls for traditional generation, storage devices and demand response. The proposed optimal controls naturally render frequency separation of these load balancing mechanisms. Given a 24-hour net load forecast, optimal tracking control inputs are designed subject to constraints of selected system elements to ensure optimal performance of the power system. Simulation results are presented and discussed.

Harvey, Roland↗

Asynchronous Navier-Stokes Solver on 3D Unstructured Grids for the Exascale Era

This project has developed multiple fluid dynamics solvers for complex 3D flows using fully asynchronous distributed-memory task-parallel algorithms on top of the Charm++ runtime system. The algorithms solve the Euler or Navier-Stokes equations of compressible flows using unstructured tetrahedron meshes with optional solution-adaptive mesh-, and polynomial-order refinement. We have demonstrated excellent strong scaling up 50K compute cores and the benefits of Charm++’s automatic load balancing.

97 MATHEMATICS AND COMPUTING↗

Advancing Grid Resilience through Smart Charge Management: Findings from Maryland’s Pilot

This report presents research findings from a four-year Smart Charge Management (SCM) pilot program conducted by Maryland’s largest electric utilities—Baltimore Gas and Electric (BGE), Potomac Electric Power Company (Pepco), and Delmarva Power & Light (DPL)—to evaluate strategies for optimizing electric vehicle (EV) charging loads and enhancing grid stability. Supported by the U.S. Department of Energy (DOE), Argonne National Laboratory collaborated with all project partners and examined the effectiveness of Time-of-Use (TOU) and Load Balancing (LB) strategies in managing peak demand, deferring costly infrastructure upgrades, and reducing grid constraints at the feeder level. Using charging data from over 4,600 EV drivers, the study analyzed SCM’s impact on the distribution systems of BGE and Pepco, which consists of over 2000 feeders. Unlike prior research that focused on system-wide trends or synthetic feeders, this analysis offers granular, feeder-level insights based on real-world operational data. It highlights how transformer density, load profiles, and infrastructure constraints influence smart charging performance. Results show feeder-level conditions play a crucial role in SCM effectiveness, with most feeders benefiting more from LB, while TOU-based SCM may be sufficient for others. By 2035, LB reduced peak charging loads by 27% on average, compared to 23% under TOU-based SCM, though some feeders saw reductions exceeding 35%, while others experienced minimal impact. Feeders with higher transformer utilization and limited capacity benefited more from LB, which more effectively distributed charging demand during off-peak hours. Beyond reducing grid constraints, SCM offers long-term operational and financial benefits. By shifting EV charging demand strategically, utilities can optimize asset utilization, delay infrastructure investments, and enhance grid performance. In terms of infrastructure upgrade deferrals, at the feeder level, LB consistently reduced peak charging loads and resulting infrastructure upgrade costs, particularly in high EV enrollment areas, decreasing the number of overloaded transformers by up to 35%, while TOU-based SCM achieved 20-30% reductions depending on feeder characteristics. At the system level, LB has the potential to defer total upgrade costs by $\$$186 million for BGE, compared to $\$$159 million under TOU-based SCM. For Pepco, TOU-based SCM performed slightly better, deferring upgrade costs by $\$$30 million, compared to $\$$29 million under LB. Section 4.5 reviews some of the system differences between BGE and Pepco. However, as EV adoption scales, TOU-based SCM will introduce secondary peak charging loads, reinforcing the need for more advanced, adaptive SCM approaches to prevent new grid challenges. As EV adoption continues to grow, feeder-level managed charging strategies will be essential for mitigating grid stress, improving infrastructure efficiency, and maintaining energy affordability for consumers. This report provides critical insights for utilities, Public Utility Commissions (PUCs), and state agencies on the role of feeder-specific smart charging in infrastructure planning, policy development, and grid modernization. The findings underscore the importance of tailored, data-driven SCM solutions that align with local grid conditions, ensuring a resilient, cost-effective transition to increasing EV adoption while safeguarding distribution system performance.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Solvers for $\mathcal{O} (N)$ Electronic Structure in the Strong Scaling Limit

Here we present a hybrid OpenMP/Charm\tt++ framework for solving the $\mathcal{O} (N)$ self-consistent-field eigenvalue problem with parallelism in the strong scaling regime, $P\gg{N}$, where $P$ is the number of cores, and $N$ is a measure of system size, i.e., the number of matrix rows/columns, basis functions, atoms, molecules, etc. This result is achieved with a nested approach to spectral projection and the sparse approximate matrix multiply [Bock and Challacombe, SIAM J. Sci. Comput., 35 (2013), pp. C72--C98], and involves a recursive, task-parallel algorithm, often employed by generalized $N$-Body solvers, to occlusion and culling of negligible products in the case of matrices with decay. Lastly, employing classic technologies associated with generalized $N$-Body solvers, including overdecomposition, recursive task parallelism, orderings that preserve locality, and persistence-based load balancing, we obtain scaling beyond hundreds of cores per molecule for small water clusters ([H${}_2$O]${}_N$, $N \in \{ 30, 90, 150 \}$, $P/N \approx \{ 819, 273, 164 \}$) and find support for an increasingly strong scalability with increasing system size $N$.

97 MATHEMATICS AND COMPUTING↗

Apparatus and method for optimal phase balancing using dynamic programming with spatial consideration

Provided are an apparatus and method for load-balancing of a three-phase electric power distribution system having a multi-phase feeder, including obtaining topology information of the feeder identifying supply points for customer loads and feeder sections between the supply points, obtaining customer information that includes peak customer load at each of the points between each of the feeder sections, performing a phase balancing analysis, and recommending phase assignment at the customer load supply points.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Pufferscale: Rescaling HPC Data Services for High Energy Physics Applications

User-space HPC data services are emerging as an appealing alternative to traditional parallel file systems, because of their ability to be tailored to application needs while eliminating unnecessary overheads incurred by POSIX compliance. The High Energy Physics (HEP) community is progressively turning towards such services to enable high-throughput accesses under heavy concurrency to billions of event data produced by instruments and consumed by subsequent analysis workflows. Such services would benefit from the possibility to be rescaled up and down to adapt to changing workloads, as experimental campaigns progress, in order to optimize resource usage. This paper formalizes rescaling a distributed storage system as a multi objective optimization problem considering three criteria: load balance, data balance, and duration of the rescaling operation. We propose a heuristic for rapidly finding a good approximate solution, while allowing users to weight the criteria as needed. The heuristic is evaluated with Pufferscale, a new rescaling manager for microservice-based distributed storage systems. To validate our approach in a real-world ecosystem, we showcase the use of Pufferscale as a means to enable storage malleability in the HEPnOS storage system for HEP applications.

Distributed Storage System↗