Introduction to Exascale Scientific Applications: Programming Approaches for Scalability, Performance, and Portability
Explore the source record for details and available documents.
SEARCH · Search NASA
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.
Explore the source record for details and available documents.
Ceph is an emerging open-source parallel distributed file and storage system. By design, Ceph leverages unreliable commodity storage and network hardware, and provides reliability and fault-tolerance via controlled object placement and data replication. This paper presents our file and block I/O performance and scalability evaluation of Ceph for scientific high-performance computing (HPC) environments. Our work makes two unique contributions. First, our evaluation is performed under a realistic setup for a large-scale capability HPC environment using a commercial high-end storage system. Second, our path of investigation, tuning efforts, and findings made direct contributions to Ceph's development and improved code quality, scalability, and performance. These changes should benefit both Ceph and the HPC community at large.
The previously established ExaCA software for performance portable alloy grain structure simulation has been updated to better represent the solidification behavior during complex alloy processing conditions, such as those encountered during metal additive manufacturing (AM), and for improved performance and scalability. Here, an extension to the time–temperature history input data format and the core ExaCA algorithm to include an arbitrary number of melting and solidification events yielded improved prediction of texture for various melt pool geometries, expanding the range of AM-relevant conditions that can be accurately simulated. Improved heat transport process simulation coupling, including the creation of large raster datasets from single track time–temperature history data and in-memory coupling with the new, performance portable finite difference code Finch, were also demonstrated in example studies on the effect of multilayer AM microstructure predictions on hatch spacing and cell size, respectively. Additional new features are detailed and demonstrated, including the ability to perform simulations using various interfacial response function forms, execute simulations on state-of-the-art hardware, improved usability through post-processing versatility, and improved strong and weak scaling performance. The performance, physics, and versatility improvements demonstrated here will further enable large-scale studies on AM process–microstructure relationships that were not previously possible. Furthermore, the usability improvements and ability to run coupled AM process–microstructure simulations using the Finch-ExaCA workflow will facilitate broader use of this open-source software by the computational materials community.
Performance tests and analyses are critical to effective high-performance computing software development and are central components in the design and implementation of computational algorithms for achieving faster simulations on existing and future computing architectures for large-scale application problems. In this article, we explore performance and space-time trade-offs for important compute-intensive kernels of large-scale numerical solvers for partial differential equations (PDEs) that govern a wide range of physical applications. We consider a sequence of PDE-motivated bake-off problems designed to establish best practices for efficient high-order simulations across a variety of codes and platforms. We measure peak performance (degrees of freedom per second) on a fixed number of nodes and identify effective code optimization strategies for each architecture. In addition to peak performance, we identify the minimum time to solution at 80% parallel efficiency. The performance analysis is based on spectral and p-type finite elements but is equally applicable to a broad spectrum of numerical PDE discretizations, including finite difference, finite volume, and h-type finite elements.
This final report contains reports of research related to the tasks "Scalable High Performance Computing: Direct and Lark-Eddy Turbulent FLow Simulations Using Massively Parallel Computers" and "Devleop High-Performance Time-Domain Computational Electromagnetics Capability for RCS Prediction, Wave Propagation in Dispersive Media, and Dual-Use Applications. The discussion of Scalable High Performance Computing reports on three objectives: validate, access scalability, and apply two parallel flow solvers for three-dimensional Navier-Stokes flows; develop and validate a high-order parallel solver for Direct Numerical Simulations (DNS) and Large Eddy Simulation (LES) problems; and Investigate and develop a high-order Reynolds averaged Navier-Stokes turbulence model. The discussion of High-Performance Time-Domain Computational Electromagnetics reports on five objectives: enhancement of an electromagnetics code (CHARGE) to be able to effectively model antenna problems; utilize lessons learned in high-order/spectral solution of swirling 3D jets to apply to solving electromagnetics project; transition a high-order fluids code, FDL3DI, to be able to solve Maxwell's Equations using compact-differencing; develop and demonstrate improved radiation absorbing boundary conditions for high-order CEM; and extend high-order CEM solver to address variable material properties. The report also contains a review of work done by the systems engineer.
Several features make Java an attractive choice for scientific applications. In order to gauge the applicability of Java to Computational Fluid Dynamics (CFD), we have implemented the NAS (NASA Advanced Supercomputing) Parallel Benchmarks in Java. The performance and scalability of the benchmarks point out the areas where improvement in Java compiler technology and in Java thread implementation would position Java closer to Fortran in the competition for scientific applications.
We address memory performance issues observed in Blue Gene Linux and discuss the design and implementation of 'Big Memory' - an alternative, transparent memory space introduced to eliminate the memory performance issues. We evaluate the performance of Big Memory using custom memory benchmarks, NAS Parallel Benchmarks, and the Parallel Ocean Program, at a scale of up to 4,096 nodes. We find that Big Memory successfully resolves the performance issues normally encountered in Blue Gene Linux. For the ocean simulation program, we even find that Linux with Big Memory provides better scalability than does the lightweight compute node kernel designed solely for high-performance applications. Originally intended exclusively for compute node tasks, our new memory subsystem dramatically improves the performance of certain I/O node applications as well. We demonstrate this performance using the central processor of the LOw Frequency ARray radio telescope as an example.
By 2009 the Fermilab Mass Storage System had encountered two major challenges: the required amount of data stored and accessed in both tiers of the system (dCache and Enstore) had significantly increased and the number of clients accessing Mass Storage System had increased from tens to hundreds of nodes and from hundreds to thousands of parallel requests. To address these challenges Enstore and the SRM part of dCache were modified to scale for performance, access rates, and capacity. This work increased the amount of simultaneously processed requests in a single Enstore Library instance from about 1000 to 30000. The rates of incoming requests to Enstore increased from tens to hundreds per second. Fermilab is invested in LTO4 tape technology and we have investigated both LTO5 and Oracle T10000C to cope with the increasing needs in capacity. We have decided to adopt T10000C, mainly due to its large capacity, which allows us to scale up the existing robotic storage space by a factor 6. This paper describes the modifications and investigations that allowed us to meet these scalability and performance challenges and provided some perspectives of Fermilab Mass Storage System.
Palacios and Kitten are new open source tools that enable applications, whether ported or not, to achieve scalable high performance on large machines. They provide a thin layer over the hardware to support both full-featured virtualized environments and native code bases. Kitten is an OS under development at Sandia that implements a lightweight kernel architecture to provide predictable behavior and increased flexibility on large machines, while also providing Linux binary compatibility. Palacios is a VMM that is under development at Northwestern University and the University of New Mexico. Palacios, which can be embedded into Kitten and other OSes, supports existing, unmodified applications and operating systems by using virtualization that leverages hardware technologies. We describe the design and implementation of both Kitten and Palacios. Our benchmarks show that they provide near native, scalable performance. Palacios and Kitten provide an incremental path to using supercomputer resources that is not performance-compromised.
In the field of additive manufacturing (AM), cellular automata (CA) is extensively used to simulate microstructural evolution during solidification. However, while traditional CA approaches are relatively fast, they still require a substantial number of time steps, are limited to moderate volumes, and are relatively difficult to improve through parallelism due to the highly localized nature of the solidification front. Here, to address these issues of time to solution and load balancing, we introduce Toucan, a parallel, performance-portable, and scalable code written in C++ with the Kokkos library that leverages the discrete event inspired cellular automata (DECA) algorithm to perform parallel-in-time (PinT) grain growth simulations. Toucan effectively mitigates load balancing issues by distributing the computational workload more evenly across processors, enhancing scalability and efficiency. We conduct both strong and weak scaling studies on up to 64 GPUs on the Frontier supercomputer, demonstrating that Toucan significantly outperforms the current state-of-the-art, time-stepped CA code, ExaCA, on both single and multi-GPU simulations. Even in AM-specific weak scaling scenarios, Toucan maintains near-ideal scaling, in contrast to the linear increase observed with ExaCA due to the moving laser raster pattern. This study highlights Toucan’s potential to transform microstructural simulations in AM by radically improving both efficiency and scalability over existing methods.
Modern experiments in high energy physics analyze millions of events recorded in particle detectors to select the events of interest and make measurements of physics parameters. These data can often be stored as tabular data in files with detector information and reconstructed quantities. Current techniques for event selection in these files lack the scalability needed for high performance computing environments. We describe our work to develop a high energy physics analysis framework suitable for high performance computing. This new framework utilizes modern tools for reading files and implicit data parallelism. Framework users analyze tabular data using standard, easy-to-use data analysis techniques in Python while the framework handles the file manipulations and parallelism without the user needing advanced experience in parallel programming. In future versions, we hope to provide a framework that can be utilized on a personal computer or a high performance computing cluster with little change to the user code.
Performance modeling is a challenging problem due to the complexities of hardware architectures. In this paper, we present PPT-GPU, a scalable and accurate simulation framework that enables GPU code developers and architects to predict the performance of applications in a fast, and accurate manner on different GPU architectures. PPT-GPU is part of the open source project, Performance Prediction Toolkit (PPT) developed at the Los Alamos National Laboratory. We extend the old GPU model in PPT that predict the runtimes of computational physics codes to offer better prediction accuracy, for which, we add models for different memory hierarchies found in GPUs and latencies for different instructions. To further show the utility of PPT-GPU, we compare our model against real GPU device(s) and the widely used cycle-accurate simulator, GPGPU-Sim using different workloads from RODINIA and Parboil benchmarks. The results indicate that the predicted performance of PPT-GPU is within a 10% error compared to the real device(s). In addition, PPT-GPU is highly scalable, where it is up to 450x faster than GPGPU-Sim with more accurate results.
Abstract On-chip spectrometers have the potential to offer dramatic size, weight, and power advantages over conventional benchtop instruments for many applications such as spectroscopic sensing, optical network performance monitoring, hyperspectral imaging, and radio-frequency spectrum analysis. Existing on-chip spectrometer designs, however, are limited in spectral channel count and signal-to-noise ratio. Here we demonstrate a transformative on-chip digital Fourier transform spectrometer that acquires high-resolution spectra via time-domain modulation of a reconfigurable Mach-Zehnder interferometer. The device, fabricated and packaged using industry-standard silicon photonics technology, claims the multiplex advantage to dramatically boost the signal-to-noise ratio and unprecedented scalability capable of addressing exponentially increasing numbers of spectral channels. We further explore and implement machine learning regularization techniques to spectrum reconstruction. Using an ‘elastic-D 1 ’ regularized regression method that we develop, we achieved significant noise suppression for both broad (>600 GHz) and narrow (<25 GHz) spectral features, as well as spectral resolution enhancement beyond the classical Rayleigh criterion.
This paper introduces PapyrusKV, a parallel embedded key-value store (KVS) for distributed high-performance computing (HPC) architectures that offer potentially massive pools of nonvolatile memory (NVM). PapyrusKV stores keys with their values in arbitrary byte arrays across multiple NVMs in a distributed system. PapyrusKV provides standard KVS operations such as put, get, and delete. More importantly, PapyrusKV provides advanced features for HPC such as dynamic consistency control, zero-copy workflow, and asynchronous checkpoint/restart. Beyond filesystems, PapyrusKV provides HPC programmers with a high-level interface to exploit distributed NVM in the system, and it transparently organizes data to achieve high performance. Also, it allows HPC applications to specialize PapyrusKV to meet their specific requirements. We empirically evaluate PapyrusKV on three HPC systems with real NVM devices: OLCF's Summitdev, TACC's Stampede, and NERSC's Cori. Our results show that PapyrusKV can offer high performance, scalability, and portability across these various distributed NVM architectures.
We report the scalable three-dimensional (3-D) integration of functional nanostructures into applicable platforms represents a promising technology to meet the ever-increasing demands of fabricating high performance devices featuring cost-effectiveness, structural sophistication and multi-functional enabling. Such an integration process generally involves a diverse array of nanostructural entities (nano-entities) consisting of dissimilar nanoscale building blocks such as nanoparticles, nanowires, and nanofilms made of metals, ceramics, or polymers. Various synthetic strategies and integration methods have enabled the successful assembly of both structurally and functionally tailored nano-arrays into a unique class of monolithic devices. The performance of nano-array based monolithic devices is dictated by a few important factors such as materials substrate selection, nanostructure composition and nano-architecture geometry. Therefore, the rational material selection and nano-entity manipulation during the nano-array integration process, aiming to exploit the advantageous characteristics of nanostructures and their ensembles, are critical steps towards bridging the design of nanostructure integrated monolithic devices with various practical applications. In this article, we highlight the latest research progress of the two-dimensional (2-D) and 3-D metal and metal oxide based nanostructural integrations into prototype devices applicable with ultrahigh efficiency, good robustness and improved functionality. Lastly, selective examples of nano-array integration, scalable nanomanufacturing and representative monolithic devices such as catalytic converters, sensors and batteries will be utilized as the connecting dots to display a roadmap from hierarchical nanostructural assembly to practical nanotechnology implications ranging from energy, environmental, to chemical and biotechnology areas.
In this paper, we propose a novel allgather algorithm, Reindexed Recursive K-ing (RRK), which leverages flexibility in the algorithm’s tree topology and ability to make asynchronous progress coupled with Core-Direct communication offload capability to optimize the MPI_Allgather for Core-Direct enabled systems. In particular, the RRK introduces a reindexing scheme which ensures contiguous data transfers while adding only a single additional send and receive operation for any radix, k, or communicator size, N. This allows us to improve algorithm scalability by avoiding the use of a scatter/gather elements (SGE) list on InfiniBand networks. The implementations of the RRK algorithm and its evaluation shows that it performs and scales well on Core-Direct systems for a wide range of message sizes and various communicator configurations.
In this paper, we propose a novel allgather algorithm, Reindexed Recursive K-ing (RRK), which leverages flexibility in the algorithm's tree topology and ability to make asynchronous progress coupled with Core-Direct communication offload capability to optimize the MPI-Allgather for Core-Direct enabled systems. In particular, the RRK introduces a reindexing scheme which ensures contiguous data transfers while adding only a single additional send and receive operation for any radix, k, or communicator size, N. This allows us to improve algorithm scalability by avoiding the use of a scatter/gather elements (SGE) list on InfiniBand networks. The implementations of the RRK algorithm and its evaluation shows that it performs and scales well on Core-Direct systems for a wide range of message sizes and various communicator configurations.
The goal of this project has been to develop a lossless compression algorithm for message-passing libraries that can accelerate HPC systems by reducing the communication time. Because both compression and decompression have to be performed in software in real time, the algorithm has to be extremely fast while still delivering a good compression ratio. During the first half of this project, they designed a new compression algorithm called FPC for scientific double-precision data, made the source code available on the web, and published two papers describing its operation, the first in the proceedings of the Data Compression Conference and the second in the IEEE Transactions on Computers. At comparable average compression ratios, this algorithm compresses and decompresses 10 to 100 times faster than BZIP2, DFCM, FSD, GZIP, and PLMI on the three architectures tested. With prediction tables that fit into the CPU's L1 data acache, FPC delivers a guaranteed throughput of six gigabits per second on a 1.6 GHz Itanium 2 system. The C source code and documentation of FPC are posted on-line and have already been downloaded hundreds of times. To evaluate FPC, they gathered 13 real-world scientific datasets from around the globe, including satellite data, crash-simulation data, and messages from HPC systems. Based on the large number of requests they received, they also made these datasets available to the community (with permission of the original sources). While FPC represents a great step forward, it soon became clear that its throughput was too slow for the emerging 10 gigabits per second networks. Hence, no speedup can be gained by including this algorithm in an MPI library. They therefore changed the aim of the second half of the project. Instead of implementing FPC in an MPI library, they refocused their efforts to develop a parallel compression algorithm to further boost the throughput. After all, all modern high-end microprocessors contain multiple CPUs on a single die and will undoubtedly contain even larger numbers of CPUs in the future. Hence, a compute node in a parallel system is likely to have multiple processors per interconnection link. Thus, they investigated how to best exploit multiple CPUs to compress and decompress data for a single link, i.e., they developed a parallel version of the FPC algorithm called pFPC. On a 4-core 3 GHz Xeon system, pFPC exceeds 10 gigabits per second both in compression and decompression throughput while still reaching a compression ratio of 18%, which is reasonable for their double-precision floating-point datasets. At half this throughput, it achieves a 41% compression ratio. Moreover, the algorithm scales well to four cores and promises even higher throughputs on larger machines. Additional results and a description of the pFPC algorithm have recently been published in the proceedings of the Data Compression Conference. The C source code of pFPC is also publicly available.