Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Parallel Supercomputers”

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 559 records · Page 31

A Many-Task Parallel Approach for Multiscale Simulations of Subsurface Flow and Reactive Transport

Continuum-scale models have long been used to study subsurface flow, transport, and reactions but lack the ability to resolve processes that are governed by pore-scale mixing. Recently, pore-scale models, which explicitly resolve individual pores and soil grains, have been developed to more accurately model pore-scale phenomena, particularly reaction processes that are controlled by local mixing. However, pore-scale models are prohibitively expensive for modeling application-scale domains. This motivates the use of a hybrid multiscale approach in which continuum- and pore-scale codes are coupled either hierarchically or concurrently within an overall simulation domain (time and space). This approach is naturally suited to an adaptive, loosely-coupled many-task methodology with three potential levels of concurrency. Each individual code (pore- and continuum-scale) can be implemented in parallel; multiple semi-independent instances of the pore-scale code are required at each time step providing a second level of concurrency; and Monte Carlo simulations of the overall system to represent uncertainty in material property distributions provide a third level of concurrency. We have developed a hybrid multiscale model of a mixing-controlled reaction in a porous medium wherein the reaction occurs only over a limited portion of the domain. Loose, minimally-invasive coupling of pre-existing parallel continuum- and pore-scale codes has been accomplished by an adaptive script-based workflow implemented in the Swift workflow system. We describe here the methods used to create the model system, adaptively control multiple coupled instances of pore- and continuum-scale simulations, and maximize the scalability of the overall system. We present results of numerical experiments conducted on NERSC supercomputing systems; our results demonstrate that loose many-task coupling provides a scalable solution for multiscale subsurface simulations with minimal overhead.

high performance computing↗

Scalable parallel distance field construction for large-scale applications

Computing distance fields is fundamental to many scientific and engineering applications. Distance fields can be used to direct analysis and reduce data. In this paper, we present a highly scalable method for computing 3D distance fields on massively parallel distributed-memory machines. Anew distributed spatial data structure, named parallel distance tree, is introduced to manage the level sets of data and facilitate surface tracking overtime, resulting in significantly reduced computation and communication costs for calculating the distance to the surface of interest from any spatial locations. Our method supports several data types and distance metrics from real-world applications. We demonstrate its efficiency and scalability on state-of-the-art supercomputers using both large-scale volume datasets and surface models. We also demonstrate in-situ distance field computation on dynamic turbulent flame surfaces for a petascale combustion simulation. In conclusion, our work greatly extends the usability of distance fields for demanding applications.

97 MATHEMATICS AND COMPUTING↗

Communication-Avoiding and Memory-Constrained Sparse Matrix-Matrix Multiplication at Extreme Scale

Sparse matrix-matrix multiplication (SpGEMM) is a widely used kernel in various graph, scientific computing and machine learning algorithms. In this paper, we consider SpGEMMs performed on hundreds of thousands of processors generating trillions of nonzeros in the output matrix. Distributed SpGEMM at this extreme scale faces two key challenges: (1) high communication cost and (2) inadequate memory to generate the output. Furthermore, we address these challenges with an integrated communication-avoiding and memory-constrained SpGEMM algorithm that scales to 262,144 cores (more than 1 million hardware threads) and can multiply sparse matrices of any size as long as inputs and a fraction of output fit in the aggregated memory. As we go from 16,384 cores to 262,144 cores on a Cray XC40 supercomputer, the new SpGEMM algorithm runs 10x faster when multiplying large-scale protein-similarity matrices.

97 MATHEMATICS AND COMPUTING↗

Institute for Sustained Performance, Energy, and Resilience (SuPER)

The University of Tennessee (UTK) and University of Texas at El Paso (UTEP) partnership supported the three main thrusts of the SUPER project---performance, energy, and resilience. The UTK-UTEP effort thus helped advance the main goal of SUPER, which was to ensure that DOE's computational scientists can successfully exploit the emerging generation of high performance computing (HPC) systems. This goal is being met by providing application scientists with strategies and tools to productively maximize performance, conserve energy, and attain resilience. The primary vehicle through which UTK provided performance measurement support to SUPER and the larger HPC community is the Performance Application Programming Interface (PAPI). PAPI is an ongoing project that provides a consistent interface and methodology for collecting hardware performance information from various hardware and software components, including most major CPUs, GPUs and accelerators, interconnects, I/O systems, and power interfaces, as well as virtual cloud environments. The PAPI software is widely used for performance modeling of scientific and engineering applications---for example, the HOMME (High Order Methods Modeling Environment) climate code, and the GAMESS and NWChem computational chemistry codes---on DOE supercomputers. PAPI is widely deployed as middleware for use by higher-level profiling, tracing, and sampling tools (e.g., CrayPat, HPCToolkit, Scalasca, Score-P, TAU, Vampir, PerfExpert), making it the de facto standard for hardware counter analysis. PAPI has established itself as fundamental software infrastructure in every application domain (spanning academia, government, and industry), where improving performance can be mission critical. Ultimately, as more application scientists migrate their applications to HPC platforms, they will benefit from the extended capabilities this grant brought to PAPI to analyze and optimize performance in these environments, whether they use PAPI directly, or via third-party performance tools. Capabilities added to PAPI through this grant include support for new architectures such as the lastest GPU and Xeon Phi accelerators, and advanced power measurement and management features. Another important topic for the UTK team was providing support for a rich ecosystem of different fault management strategies in the context of parallel computing. Our long term efforts have been oriented toward proposing flexible strategies and providing building boxes that application developers can use to build the most efficient fault management technique for their application. These efforts span across the entire software spectrum, from theoretical models of existing strategies to easily assess their performance, to algorithmic modifications to take advantage of specific mathematical properties for data redundancy and to extensions to widely used programming paradigms to empower the application developers to deal with all types of faults. We have also continued our tight collaborations with users to help them adopt these technologies to ensure their application always deliver meaningful scientific data. Large supercomputer systems are becoming more and more power and energy constrained, and future systems and applications running on them will need to be optimized to run under power caps and/or minimize energy consumption. The UTEP team contributed to the SUPER energy thrust by developing power modeling methodologies and investigating power management strategies. Scalability modeling results showed that some applications can scale better with respect to an increasing power budget than with respect to only the number of processors. Power management, in particular shifting power to processors on the critical path of an application execution, can reduce perturbation due to system noise and other sources of runtime variability, which are growing problems on large-scale power-constrained computer systems.

42 ENGINEERING↗

Frontier (HPE Cray EX) Exascale Supercomputer at the Oak Ridge Leadership Computing Facility

Frontier is the HPE Cray EX exascale supercomputer deployed and operated by the Oak Ridge Leadership Computing Facility (OLCF) at Oak Ridge National Laboratory (ORNL). Frontier is designed for large-scale modeling, simulation, and AI workloads and is built from HPE Cray EX system architecture with AMD CPUs and AMD Instinct GPU accelerators connected by the HPE Slingshot interconnect. System composition (representative production configuration): Frontier is composed of approximately 74 cabinets with 128 compute nodes per cabinet (~9,400 compute nodes total). Each compute node contains one 64-core AMD EPYC CPU and four AMD Instinct MI250X GPUs. Nodes are connected using HPE Slingshot (Slingshot-200 class) networking with multiple NIC ports per node providing high injection bandwidth. Frontier is connected to the Orion parallel file system (multi-tier Lustre) providing a large, center-wide high-performance storage namespace. Operational context: Frontier entered public prominence as the first system to reach No. 1 on the TOP500 list in May 2022 (HPL benchmark), establishing the first widely recognized exascale-era performance milestone. The system supports DOE Office of Science mission workloads and enables leadership-class computational science and AI for open science users.

AMD EPYC↗

The Practical Obstacles of Data Transfer: Why researchers still love scp

The importance of computing facilities is heralded every six months with the announcement of the new Top500 list, showcasing the world s fastest supercomputers. Unfortu- nately, with great computing capability does not come great long-term data storage capacity, which often means users must move their data to their local site archive, to remote sites where they may be doing future computation or anal- ysis, or back to their home institution, else face the dreaded data purge that most HPC centers employ to keep utiliza- tion of large parallel filesystems low to manage performance and capacity. At HPC centers, data transfer is crucial to the scientific workflow and will increase in importance as computing systems grow in size. The Energy Sciences Net- work (ESnet) recently launched its fifth generation network, a 100 Gbps high-performance, unclassified national network connecting more than 40 DOE research sites to support scientific research and collaboration. Despite the tenfold increase in bandwidth to DOE research sites amenable to multiple data transfer streams and high throughput, in prac- tice, researchers often under-utilize the network and resort to painfully-slow single stream transfer methods such as scp to avoid the complexity of using multiple stream tools such as GridFTP and bbcp, and contend with frustration from the lack of consistency of available tools between sites. In this study we survey and assess the data transfer methods pro- vided at several DOE supported computing facilities, includ- ing both leadership-computing facilities, connected through ESnet. We present observed transfer rates, suggested opti- mizations, and discuss the obstacles the tools must overcome to receive wide-spread adoption over scp.

Nam, Hai Ah↗

Demonstrating Cross-Facility Data Processing at Scale with Laue Microdiffraction

In February and April 2023 live, at-scale data processing demonstrations were conducted between the Advanced Photon Source (APS), a synchrotron light source, and the Argonne Leadership Computing Facility (ALCF). These tests were run as part of a novel beamline technique: coded aperture laue micro-diffraction. This technique requires a significant amount of compute to decode appeture patterns embedded in the detector stream. An autonomous system was able to send data to ALCF during an experiment, utilize 50 nodes of the Polaris supercomputer to process 6-12 hour scans, and return the data back to the APS within 12-15 minutes behind the detector. With scan points arriving every 72 seconds, the system kept up with the beamline, potentially enabling in-experiment analysis. The data processing system utilizes Globus infrastructure and an on-demand queue to dynamically acquire nodes on Polaris. The underlying reconstruction algorithms were parallelized via MPI and accelerated with custom CUDA kernels.

Prince, Michael↗

Orchestrating Fault Prediction with Live Migration and Checkpointing

Checkpoint/Restart (C/R) is widely used to provide fault tolerance on High-Performance Computing (HPC) systems. However, Parallel File System (PFS) overhead and failure uncertainty cause significant application overhead. This paper develops an adaptive multi-level C/R model that incorporates a failure prediction and analysis model, which orchestrates failure prediction, checkpointing, checkpoint frequency, and proactive live migration along with the additional benefit of Burst Buffers (BB). It effectively reduces the overheads due to failures, checkpointing, and recovery. Simulation results for the Summit supercomputer yield a reduction of ~20%-86% in application overhead due to BBs, orchestrated failure prediction, and migration. We also observe a ~29% decrease in checkpoint writes to BBs, which can increase the longevity of the BB storage devices.

Behera, Subhendu↗

Using Compiler Directives for Performance Portability in Scientific Computing: Kernels from Molecular Simulation

Achieving performance portability for high-performance computing (HPC) applications in scientific fields has become an increasingly important initiative due to large differences in emerging supercomputer architectures. Here we test some key kernels from molecular dynamics (MD) to determine whether the use of the OpenACC directive-based programming model when applied to these kernels can result in performance within an acceptable range for these types of programs in the HPC setting. We find that for easily parallelizable kernels, performance on the GPU remains within this range. On the CPU, OpenACC-parallelized pairwise distance kernels would not meet the performance standards required, when using AMD Opteron “Interlagos” processors, but with IBM Power 9 processors, performance remains within an acceptable range for small batch sizes. These kernels provide a test for achieving performance portability with compiler directives for problems with memory-intensive components as are often found in scientific applications.

Sedova, Ada↗

Immortal rays: Rethinking random ray neutron transport on GPU architectures

The Random Ray Method (TRRM) is a recently developed adaptation of the Method of Characteristics for neutral particle transport simulations. TRRM has demonstrated excellent performance on 3D nuclear reactor benchmark problems using CPU-based compute systems. When porting to GPU-based systems, however, new performance challenges arise that are unique to processors targeting massive fine-grained parallelism. For smaller problems, or for large problems that are domain decomposed across many computational nodes, the problem size per node has insufficient parallelism to saturate GPU node resources, thus greatly limiting speedup. In this study, we report on a newly developed “immortal ray” variant of TRRM. Here, the immortal ray technique exposes significantly more fine-grained parallelism by fundamentally reformulating the numerical details of ray discretization, resulting in performance tradeoffs with significant overall benefit on GPUs. For very small 2D simulation problems we found the new immortal ray variant allowed for up to a 4.4x speedup when run on a single GPU. For larger 3D simulation problems we found the new variant improved strong scaling by 3x when run on the Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗

Lustre Unveiled: Evolution, Design, Advancements, and Current Trends

The Lustre filesystem serves as a vital element in high-performance parallel storage, meeting the rising demands of scientific, research, and enterprise environments. Widely deployed across HPC environments, ranging from small-scale applications in AI/ML, to domains like oil and gas, drug discovery, and meteorology, and manufacturing, Lustre addresses the universal challenge of efficiently accessing vast and ever-increasing volumes of data. Lustre is the filesystem of choice on six out of the top 10 fastest supercomputers in the world today, over 65% of the top 100, and also for over 60% of the top 500. Despite its widespread popularity, there is a lack of a complete and up-to-date reference, covering Lustre’s evolution, design, and various advancements made over the years. In this journal, we aim to fill this gap by providing a comprehensive journey of Lustre, including its history with significant contributions to HPC, detailed architecture and design elements, exploration of advancements added through its evolution, and future directions. Additionally, we present a comparison of Lustre with other prominent storage technologies of the era. To illustrate the current state of Lustre, we analyze several filesystem trends, including utilization, performance, and usage patterns on Orion, the Lustre filesystem on the first exascale supercomputer Frontier. We hope that this journal serves as a comprehensive educational reference for the current and future generations interested in HPC filesystem storage aspects.

97 MATHEMATICS AND COMPUTING↗

Dense LU Factorization on Multicore Supercomputer Nodes

Dense LU factorization is a prominent benchmark used to rank the performance of supercomputers. Many implementations, including the reference code HPL, use block-cyclic distributions of matrix blocks onto a two-dimensional process grid. The process grid dimensions drive a trade-off between communication and computation and are architecture- and implementation-sensitive. We show how the critical panel factorization steps can be made less communication-bound by overlapping asynchronous collectives for pivot identification and exchange with the computation of rank-k updates. By shifting this trade-off, a modified block-cyclic distribution can beneficially exploit more available parallelism on the critical path, and reduce panel factorization's memory hierarchy contention on now-ubiquitous multi-core architectures. The missed parallelism in traditional block-cyclic distributions arises because active panel factorization, triangular solves, and subsequent broadcasts are spread over single process columns or rows (respectively) of the process grid. Increasing one dimension of the process grid decreases the number of distinct processes in the other dimension. To increase parallelism in both dimensions, periodic 'rotation' is applied to the process grid to recover the row-parallelism lost by a tall process grid. During active panel factorization, rank-1 updates stream through memory with minimal reuse. In a column-major process grid, the performance of this access pattern degrades as too many streaming processors contend for access to memory. A block-cyclic mapping in the more popular row-major order does not encounter this problem, but consequently sacrifices node and network locality in the critical pivoting steps. We introduce 'striding' to vary between the two extremes of row- and column-major process grids. As a test-bed for further mapping experiments, we describe a dense LU implementation that allows a block distribution to be defined as a general function of block to processor. Other mappings can be tested with only small, local changes to the code.

99 GENERAL AND MISCELLANEOUS↗

UPC++: A PGAS extension for C++

Partitioned Global Address Space (PGAS) languages are convenient for expressing algorithms with large, random-access data, and they have proven to provide high performance and scalability through lightweight one-sided communication and locality control. While very convenient for moving data around the system, PGAS languages have taken different views on the model of computation, with the static Single Program Multiple Data (SPMD) model providing the best scalability. In this paper we present UPC++, a PGAS extension for C++ that has three main objectives: 1) to provide an object-oriented PGAS programming model in the context of the popular C++ language, 2) to add useful parallel programming idioms unavailable in UPC, such as asynchronous remote function invocation and multidimensional arrays, to support complex scientific applications, 3) to offer an easy on-ramp to PGAS programming through interoperability with other existing parallel programming systems (e.g., MPI, OpenMP, CUDA). We implement UPC++ with a 'compiler-free' approach using C++ templates and runtime libraries. We borrow heavily from previous PGAS languages and describe the design decisions that led to this particular set of language features, providing significantly more expressiveness than UPC with very similar performance characteristics. We evaluate the programmability and performance of UPC++ using five benchmarks on two representative supercomputers, demonstrating that UPC++ can deliver excellent performance at large scale up to 32K cores while offering PGAS productivity features to C++ applications.

Zheng, Y↗

Livermore tomography tools: Accurate, fast, and flexible software for tomographic science

Livermore Tomography Tools (LTT) is a customizable scientific software package that enables a broad range of research and development efforts into computed tomography (CT). Here, it was developed to process x-ray and neutron CT data accurately and rapidly from raw detector counts to reconstructed volumes with the flexibility to handle many special cases. LTT fulfills long-term CT software goals to provide quantitatively accurate results reported in physical units (e.g., mm -1 or cm -1 ) while exploiting all available computational advantages to maximize speed. Written in C/C++ with support for multiple CPUs and GPUs, LTT runs on many computing platforms (Linux/Unix, Windows, and Mac; laptops to supercomputers). As a result, LTT can:process data acquired from various custom-built and commercially available CT scanners, model and simulate x-ray and neutron interactions to encourage algorithm prototyping, and allow for rapid insertion of the latest algorithms.We describe LTT’s software architecture, user interfaces, and its 88 algorithms (as of this writing) for pre-processing, reconstruction, post-processing, and simulation that support many scanner geometries (parallel-, fan-, cone-beam, and custom). Several applications are presented that illustrate LTT’s accuracy, speed, and flexibility relative to other solutions.

36 MATERIALS SCIENCE↗

Characterization and identification of HPC applications at leadership computing facility

High Performance Computing (HPC) is an important method for scientific discovery via large-scale simulation, data analysis, or artificial intelligence. Leadership-class supercomputers are expensive, but essential to run large HPC applications. The Petascale era of supercomputers began in 2008, with the first machines achieving performance in excess of one petaflops, and with the advent of new supercomputers in 2021 (e.g., Aurora, Frontier), the Exascale era will soon begin. However, the high theoretical computing capability (i.e., peak FLOPS) of a machine is not the only meaningful target when designing a supercomputer, as the resources demand of applications varies. A deep understanding of the characterization of applications that run on a leadership supercomputer is one of the most important ways for planning its design, development and operation. In order to improve our understanding of HPC applications, user demands and resource usage characteristics, we perform correlative analysis of various logs for different subsystems of a leadership supercomputer. This analysis reveals surprising, sometimes counter-intuitive patterns, which, in some cases, conflicts with existing assumptions, and have important implications for future system designs as well as supercomputer operations. For example, our analysis shows that while the applications spend significant time on MPI, most applications spend very little time on file I/O. Combined analysis of hardware event logs and task failure logs show that the probability of a hardware FATAL event causing task failure is low. Combined analysis of control system logs and file I/O logs reveals that pure POSIX I/O is used more widely than higher level parallel I/O. Based on holistic insights of the application gained through combined and co-analysis of multiple logs from different perspectives and general intuition, we engineer features to "fingerprint" HPC applications. We use t-SNE (a machine learning technique for dimensionality reduction) to validate the explainability of our features and finally train machine learning models to identify HPC applications or group those with similar characteristic. To the best of our knowledge, this is the first work that combines logs on file I/O, computing, and inter-node communication for insightful analysis of HPC applications in production.

Liu, Zhengchun↗

Final Technical Report for Grant No. DE-FC02-01ER54648, SciDAC Center for Simulation of Wave-Plasma Interactions

Harnessing the energy that is released in fusion reactions would provide a safe and abundant source of power to meet the growing energy needs of the world population. The next step toward the development of fusion as a practical energy source is the construction of ITER, a device capable of producing and controlling the high performance plasma required for self-sustaining fusion reactions, or “burning” plasma. The input power required to drive the ITER plasma into the burning regime will be supplied primarily with a combination of external power from radio frequency waves in the ion cyclotron range of frequencies and energetic ions from neutral beam injection sources, in addition to internally-generated Ohmic heating from the induced plasma current that also serves to create the magnetic equilibrium for the discharge. Since the success of the ITER project depends critically on the ability to create and maintain burning plasma conditions, it is absolutely necessary to have physics-based models that can accurately simulate the RF processes that affect the dynamical evolution of the ITER discharge. The Center for Simulation of Wave-Plasma Interactions (CSWPI) is a multi-institutional collaboration that has conducted ongoing research aimed at developing: (1) Coupled core-to-edge simulations that will lead to an increased understanding of parasitic losses of the applied RF power in the boundary plasma between the RF antenna and the core plasma; (2) Development of models for core interactions of RF waves with energetic electrons and ions (including fusion alpha particles and fast neutral beam ions) that include a more accurate representation of the particle dynamics in the combined equilibrium and wave fields; and (3) Development of improved algorithms that will take advantage of massively parallel computing platforms at the petascale level and beyond to achieve the needed physics, resolution, and/or statistics to address these issues. This research reported in this final technical report is focused on further verification and validation of simulation models for interactions of RF waves with energetic particles in the core with a long term goal of assessing their readiness for inclusion in integrated modeling frameworks and utilizing the world’s largest supercomputers to quantitatively predict the detailed electric field structure in the vicinity of the launching structures.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗