Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “high performance and distributed computing”

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 109 records · Page 6

Final Report for DOE Award ER25756

The SciDAC-funded Center for Enabling Distributed Petascale Science (CEDPS) was established to address technical challenges that arise due to the frequent geographic distribution of data producers (in particular, supercomputers and scientific instruments) and data consumers (people and computers) within the DOE laboratory system. Its goal is to produce technical innovations that meet DOE end-user needs for (a) rapid and dependable placement of large quantities of data within a distributed high-performance environment, and (b) the convenient construction of scalable science services that provide for the reliable and high-performance processing of computation and data analysis requests from many remote clients. The Center is also addressing (c) the important problem of troubleshooting these and other related ultra-high-performance distributed activities from the perspective of both performance and functionality

97 MATHEMATICS AND COMPUTING↗

MeDICi: An Open Platform for Sensor Integration

MeDICi (Middleware for Data Intensive Computing) is a platform for developing high performance, distributed streaming analytic and scientific applications. Developed at Pacific Northwest National Laboratory (PNNL), MeDICi has been released under an open source license and is based on enterprise-proven middleware technologies including a widely used Enterprise Service Bus (ESB), the standard Business Process Execution Language (BPEL), and open source message brokers. Wherever possible, we have built on existing open source, standards-based systems and integrated them into a coherent whole by creating simplified graphical programming tools such as a Workflow Designer and an easy to use and well-documented integration API. This software development approach allows us to: avoid re-creating complex service integration and orchestration systems, reap the benefits of continual improvements to the technology base, and focus on creating tools and APIs which allow for the creation of re-usable component-based software components applications and workflows. These aspects have facilitated rapid adoption of the platform within PNNL for demonstration and operational applications. In fact, MeDICi has been used for a wide range of integration projects including two sensor integration applications described later on in this paper. The remainder of this article white paper is organized as follows: Section 2 provides a high-level description of the MeDICi architecture. In Section 3, the open aspects of the API and tool development are highlighted. Section 4 explains system readiness by presenting relevant demonstrations and deployments. Finally documentation and licensing details are provided in Section 5

99 GENERAL AND MISCELLANEOUS↗

LATTE: open-source, high-performance traveltime computation, tomography and source location in acoustic and elastic media

Traveltime-based tomography and source location are fundamental approaches for imaging subsurface structures and understanding the spatiotemporal distribution of seismicity from local to global scales. We present an open-source, high-performance framework integrating eikonal equation solvers and adjoint-state theory for traveltime computation, velocity tomography, source location and joint tomography-location in 2-D/3-D acoustic and elastic media. We introduce novel regularization schemes based on total generalized p-variation, structural similarity and multitask machine learning to enhance the fidelity and interpretability of inverted models and source locations. Key features of our implementation also include the ability to leverage both absolute-difference and double-difference traveltime misfits for high-fidelity velocity tomography and source parameter estimation; support for traveltime computation and inversion in diverse 2-D/3-D scenarios with arbitrary source and receiver distributions; and a perturbation-based optimal step-size estimation method to reduce computational costs. In addition, our implementation employs shared-memory and distributed-memory parallelization to provide an efficient solution for traveltime computation, tomography, and source location. In conclusion, we validate the efficacy and accuracy of our approach through multiple synthetic data examples.

58 GEOSCIENCES↗

Design and Implementation of a High-Performance CCA Event Service

Event services based on publish-subscribe architectures are well established components of distributed computing applications. Recently, an event service has been proposed as part of the Common Component Architecture (CCA) for high-performance computing applications. In this paper we describe our implementation, experimental evaluation, and initial experience with a high performance CCA event service that exploits efficient communications mechanisms commonly used on high-performance computing (HPC) platforms. We describe the CCA event service model, and discuss briefly the possible implementation strategies of that model. We then present the design and implementation of the event service using the Aggregate Remote Memory Copy Interface (ARMCI) as an underlying communication layer for this mechanism. Two alternative implementations are presented and evaluated on a Cray XD-1 platform. The performance results demonstrate that event delivery latencies are low and that the event service is able to achieve high throughput levels. Finally, we describe the use of the event service in an application for high-speed processing of data from a mass spectrometer, and conclude by discussing some possible extensions to the event service for other HPC applications.

97 MATHEMATICS AND COMPUTING↗

Data transfer for STAR grid jobs

The Solenoidal Tracker at RHIC (STAR) is a multipurpose experiment at the Relativistic Heavy Ion Collider (RHIC) with the primary goal to study the formation and properties of the quark-gluon plasma. STAR is an international collaboration of member institutions and laboratories from around the world. Yearly data-taking period produces PBytes of raw data collected by the experiment. STAR primarily uses its dedicated facility at BNL to process this data, but has routinely leveraged distributed systems, both high throughput (HTC) and high performance (HPC) computing clusters, to significantly augment the processing capacity available to the experiment. The ability to automate the efficient transfer of large data sets on reliable, scalable, and secure infrastructure is critical for any large-scale distributed processing campaign. For more than a decade, STAR computing has relied upon GridFTP with its x509-based authentication to build such data transfer systems and integrate them into its larger production workflow. The end of support by the community for both GridFTP and the x509 standard requires STAR to investigate other approaches to meet its distributed processing needs. In this study we investigate two multi-purpose data distribution systems, Globus.org and XRootD, as alternatives to GridFTP. We compare both their performance and the ease by which each service is integrated into the type of secure and automated data transfer systems STAR has previously built using GridFTP. The presented approach and study may be applicable to other distributed data processing use cases beyond STAR.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Iterative Importance Sampling Algorithms for Parameter Estimation

In parameter estimation problems one computes a posterior distribution over uncertain parameters defined jointly by a prior distribution, a model, and noisy data. Markov chain Monte Carlo (MCMC) is often used for the numerical solution of such problems. An alternative to MCMC is importance sampling, which can exhibit near perfect scaling with the number of cores on high performance computing systems because samples are drawn independently. However, finding a suitable proposal distribution is a challenging task. Several sampling algorithms have been proposed over the past years that take an iterative approach to constructing a proposal distribution. We investigate the applicability of such algorithms by applying them to two realistic and challenging test problems, one in subsurface flow, and one in combustion modeling. More specifically, we implement importance sampling algorithms that iterate over the mean and covariance matrix of Gaussian or multivariate t-proposal distributions. Our implementation leverages massively parallel computers, and we present strategies to initialize the iterations using “coarse” MCMC runs or Gaussian mixture models.

97 MATHEMATICS AND COMPUTING↗

Scalable Proximity-Based Methods for Large-Scale Analysis of Atom Probe Data

Powered by recent advances in data acquisition technologies, today's state-of-the-art atom probe microscopes yield data sets with sizes ranging from a few million atoms to billions of atoms. Analysis of these atomic data sets within rea-sonable turnaround times is a pressing data analysis challenge for material scientists currently equipped with software systems that do not scale to these massive data sets. Here, we present the shared memory component of a larger ongoing effort to develop a multi-feature data analysis framework capable of analyzing atom probe data of all sizes and scales from desktop multicore machines to large-scale high-performance computing platforms with hybrid (shared and distributed memory) architectures. Our focus here is on a broad class of popular atom probe data analysis methods that rely on core time-consuming k-NN queries. We present a scalable, heuristic algorithm for k-NN queries using three-dimensional range trees. To demonstrate its efficacy, the k-NN algorithm is integrated with two use cases of atom probe data analysis methods and the resulting analysis times are shown to speedup by over 20X on a 32-core Cray XC40 node using workloads up to 8 million atoms, which is already beyond the at-scale capabilities of existing atom probe software. Using this k-NN algorithm, we also introduce a novel parameter estimation method for a class of cluster finding methods, called friends-of-friends (FoF) methods, to completely bypass their expensive pre-processing steps. In each case, we validate the results on a variety of control data sets.

Lu, Hao↗

Secure Federated Learning Across Heterogeneous Cloud and High-Performance Computing Resources: A Case Study on Federated Fine-Tuning of LLaMA 2

Federated learning enables multiple data owners to collaboratively train robust machine learning models without transferring large or sensitive local datasets by only sharing the parameters of the locally trained models. Here, in this article, we elaborate on the design of our Advanced Privacy-Preserving Federated Learning (APPFL) framework, which streamlines end-to-end secure and reliable federated learning experiments across cloud computing facilities and high-performance computing resources by leveraging Globus Compute, a distributed function as a service platform, and Amazon Web Services. We further demonstrate the use case of APPFL in fine-tuning an LLaMA 2 7B model using several cloud resources and supercomputers.

97 MATHEMATICS AND COMPUTING↗

A high-performance parallel algorithm for nonnegative matrix factorization

Non-negative matrix factorization (NMF) is the problem of determining two non-negative low rank factors W and H, for the given input matrix A, such that A ≈ WH. NMF is a useful tool for many applications in different domains such as topic modeling in text mining, background separation in video analysis, and community detection in social networks. Despite its popularity in the data mining community, there is a lack of efficient distributed algorithms to solve the problem for big data sets. We propose a high-performance distributed-memory parallel algorithm that computes the factorization by iteratively solving alternating non-negative least squares (NLS) subproblems for W and H. It maintains the data and factor matrices in memory (distributed across processors), uses MPI for interprocessor communication, and, in the dense case, provably minimizes communication costs (under mild assumptions). As opposed to previous implementations, our algorithm is also flexible: (1) it performs well for both dense and sparse matrices, and (2) it allows the user to choose any one of the multiple algorithms for solving the updates to low rank factors W and H within the alternating iterations. We demonstrate the scalability of our algorithm and compare it with baseline implementations, showing significant performance improvements.

Kannan, Ramakrishnan↗

xGFabric: Coupling Sensor Networks and HPC Facilities with Private 5G Wireless Networks for Real-Time Digital Agriculture

Advanced scientific applications require coupling distributed sensor networks with centralized high-performance computing facilities. Citrus Under Protective Screening (CUPS) exemplifies this need in digital agriculture, where citrus research facilities are instrumented with numerous sensors monitoring environmental conditions and detecting protective screening damage. CUPS demands access to computational fluid dynamics codes for modeling environmental conditions and guiding real-time interventions like water application or robotic repairs. These computing domains have contrasting properties: sensor networks provide low-performance, limited-capacity, unreliable data access, while high-performance facilities offer enormous computing power through high-latency batch processing. Private 5G networks present novel capabilities addressing this challenge by providing low latency, high throughput, and reliability necessary for near-real-time coupling of edge sensor networks with HPC simulations. This work presents xGFabric, an end-to-end system coupling sensor networks with HPC facilities through Private 5G networks. The prototype connects remote sensors via 5G network slicing to HPC systems, enabling real-time digital agriculture simulation.

Digital Agriculture↗

HPC-NMF: A High-Performance Parallel Algorithm for Nonnegative Matrix Factorization

NMF is a useful tool for many applications in different domains such as topic modeling in text mining, background separation in video analysis, and community detection in social networks. Despite its popularity in the data mining community, there is a lack of efficient distributed algorithms to solve the problem for big data sets. We propose a high-performance distributed-memory parallel algorithm that computes the factorization by iteratively solving alternating non-negative least squares (NLS) subproblems for $\WW$ and $\HH$. It maintains the data and factor matrices in memory (distributed across processors), uses MPI for interprocessor communication, and, in the dense case, provably minimizes communication costs (under mild assumptions). As opposed to previous implementation, our algorithm is also flexible: It performs well for both dense and sparse matrices, and allows the user to choose any one of the multiple algorithms for solving the updates to low rank factors $\WW$ and $\HH$ within the alternating iterations.

Kannan, Ramakrishnan↗

A Case Study of MPI Over Long Distance Connections

Scientific workflows are increasingly being distributed across wide-area networks, and their code executions are expected to span across geographically dispersed computing systems. MPI has been extensively used to support communications for distributed computations, typically, over compute clusters and high-performance systems within a single facility. We present a case study of performance of MPI basic operations over long distance connections, wherein TCP is used for the underlying transport. We present measurements of execution times of MPI codes that utilize MPI Sendrecv operations over emulated 10Gbps connections with 0-366ms round-trip times, including the longest one spanning the globe. They demonstrate that basic MPI codes can be sustained over long distance connections under external packet loss rates up to 10%. They also highlight the qualitative effects of losses which manifest as increased execution times as a consequence of TCP’s loss recovery process.

Rao, Nageswara S.↗

Integration of sensing and computing in an intelligent decision support system for homeland security defense

We propose an intelligent decision support system based on sensor and computer networks that incorporates various component techniques for sensor deployment, data routing, distributed computing, and information fusion. The integrated system is deployed in a distributed environment composed of both wireless sensor networks for data collection and wired computer networks for data processing in support of homeland security defense. We present the system framework and formulate the analytical problems and develop approximate or exact solutions for the subtasks: (i) sensor deployment strategy based on a two-dimensional genetic algorithm to achieve maximum coverage with cost constraints; (ii) data routing scheme to achieve maximum signal strength with minimum path loss, high energy efficiency, and effective fault tolerance; (iii) network mapping method to assign computing modules to network nodes for high-performance distributed data processing; and (iv) binary decision fusion rule that derive threshold bounds to improve system hit rate and false alarm rate. These component solutions are implemented and evaluated through either experiments or simulations in various application scenarios. The extensive results demonstrate that these component solutions imbue the integrated system with the desirable and useful quality of intelligence in decision making.

59 BASIC BIOLOGICAL SCIENCES↗

Pele: An Exascale-Ready Suite of Combustion Codes

High fidelity simulations of realistic combustion devices are extremely demanding computationally because of the requirements to capture complex fuel chemical decomposition, its intricate interactions with turbulent, often multiphase, flows, and the wide separation of space and time scales between the thin flame and the device boundaries. Software required to carry out such computations tends to be extremely complex, particularly when designed to exploit hardware accelerators, and can be difficult to port and maintain. We present Pele, a performance portable suite of tools for the simulation of combustion systems, including codes to evolve reactive multiphase configurations in the low Mach number and compressible flow regimes, along with a set of inter-compatible post processing and in situ analysis tools. The Pele suite of tools is built on top of the AMReX framework for block-structured adaptive mesh refinement, which provides efficient data structures and algorithms that enable the development of a wide variety of efficient mesh and particle based PDE integration schemes. A hierarchical MPI+X parallelism scheme supports CPU-only and accelerated architectures, where X can be OpenMP, CUDA, and HIP based approaches for intra-node computational work distribution. The algorithms and data structures underlying the Pele simulation and analysis tools are highly scalable and performant across a wide variety of high-performance computing platforms, including DOEs newest exascale-class machines, Frontier and Aurora. The simulation and analysis tools are fully documented and freely distributed as open source via GitHub. We present key algorithmic and software challenges, solution strategies, performance and resulting set of capabilities.

AMReX↗

Distributed deep learning training using silicon photonic switched architectures

The scaling trends of deep learning models and distributed training workloads are challenging network capacities in today’s datacenters and high-performance computing (HPC) systems. We propose a system architecture that leverages silicon photonic (SiP) switch-enabled server regrouping using bandwidth steering to tackle the challenges and accelerate distributed deep learning training. In addition, our proposed system architecture utilizes a highly integrated operating system-based SiP switch control scheme to reduce implementation complexity. To demonstrate the feasibility of our proposal, we built an experimental testbed with a SiP switch-enabled reconfigurable fat tree topology and evaluated the network performance of distributed ring all-reduce and parameter server workloads. The experimental results show up to 3.6× improvements over the static non-reconfigurable fat tree. Our large-scale simulation results show that server regrouping can deliver up to 2.3× flow throughput improvement for a 2× tapered fat tree and a further 11% improvement when higher-layer bandwidth steering is employed. The collective results show the potential of integrating SiP switches into datacenters and HPC systems to accelerate distributed deep learning training.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗