Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel systems”

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 217 records · Page 12

Rethinking key–value store for parallel I/O optimization

Key-value stores are being widely used as the storage system for large-scale internet services and cloud storage systems. However, they are rarely used in HPC systems, where parallel file systems are the dominant storage solution. In this study, we examine the architecture differences and performance characteristics of parallel file systems and key-value stores. We propose using key-value stores to optimize overall Input/Output (I/O) performance, especially for workloads that parallel file systems cannot handle well, such as the cases with intense data synchronization or heavy metadata operations. We conducted experiments with several synthetic benchmarks, an I/O benchmark, and a real application. We modeled the performance of these two systems using collected data from our experiments, and we provide a predictive method to identify which system offers better I/O performance given a specific workload. The results show that we can optimize the I/O performance in HPC systems by utilizing key-value stores.

Hyperdex↗

Fundamental Properties of Low-Mass Stars and Brown Dwarfs

Precise measurements of the fundamental properties of low-mass stars and brown dwarfs are key to understanding the physics underlying their formation and evolution. While there has been great progress over the last decade in studying the bulk spectrophotometric properties of low-mass objects, direct determination of their masses, radii, and temperatures have been very sparse. Thus, theoretical predictions of low-mass evolution and ultracool atmospheres remain to be rigorously tested. The situation is alarming given that such models are widely used, from the determination of the low-mass end of the initial mass function to the characterization of exoplanets.An increasing number of mass, radius, and age determinations are placing critical constraints on the physics of low-mass objects. A wide variety of approaches are being pursued, including eclipsing binary studies, astrometric-spectroscopic orbital solutions, interferometry, and characterization of benchmark systems. In parallel, many more systems suitable for concerted study are now being found, thanks to new capabilities spanning both the very widest (all-sky surveys) and very narrowest (diffraction-limited adaptive optics) areas of the sky. This Cool Stars 15 splinter session highlighted the current successes and limitations of this rapidly growing area of precision astrophysics.

39↗

FitCache: A Transparent Drop-In Framework for Multi-Tier Caching to Accelerate Distributed Deep Learning Workloads

Training in Deep learning (DL) remains highly compute- and data-intensive, with I/O becoming a critical bottleneck as models and datasets scale. Recent studies report that data loading can dominate training time, especially on large-scale HPC systems with shared parallel file systems (PFS). Existing caching approaches either rely on single-tier designs or require intrusive modifications to training pipelines, limiting their portability and effectiveness. In this work, we present FitCache, a transparent drop-in framework for multi-tier caching to accelerate distributed DL training by coordinating fast local memory (e.g., DRAM, Persistent Memory (PMem)) and NVMe as hierarchical caches atop PFS. Our design adapts to hardware diversity, i.e., if NVMe is missing, memory transparently acts as a caching tier, ensuring stable performance. FitCache transparently intercepts I/O requests and issues concurrent fetches across all tiers, returning data from the fastest responder without centralized metadata or static redirection paths. FitCache adapts to dynamic workloads and heterogeneous clusters while maintaining POSIX compatibility. Experiments on Frontier (2048 GPUs) and smaller research clusters show that FitCache reduces training time by up to 40% and per-batch I/O latency by up to 71.6% compared to Lustre Orion PFS, offering a drop-in solution for scalable DL training.

Hu, Guangxing [ORNL] (ORCID:0009000283203614)↗

Development of a pixelated GSO gamma camera system with tungsten parallel hole collimator for single photon imaging

Purpose: In small animal imaging using a single photon emitting radionuclide, a high resolution gamma camera is required. Recently, position sensitive photomultiplier tubes (PSPMTs) with high quantum efficiency have been developed. By combining these with nonhygroscopic scintillators with a relatively low light output, a high resolution gamma camera can become useful for low energy gamma photons. Therefore, the authors developed a gamma camera by combining a pixelated Ce-doped Gd{sub 2}SiO{sub 5} (GSO) block with a high quantum efficiency PSPMT. Methods: GSO was selected for the scintillator, because it is not hygroscopic and does not contain any natural radioactivity. An array of 1.9 mm x 1.9 mm x 7 mm individual GSO crystal elements was constructed. These GSOs were combined with a 0.1-mm thick reflector to form a 22 x 22 matrix and optically coupled to a high quantum efficiency PSPMT (H8500C-100 MOD8). The GSO gamma camera was encased in a tungsten gamma-ray shield with tungsten pixelated parallel hole collimator, and the basic performance was measured for Co-57 gamma photons (122 keV). Results: In a two-dimensional position histogram, all pixels were clearly resolved. The energy resolution was {approx}15% FWHM. With the 20-mm thick tungsten pixelated collimator, the spatial resolution was 4.4-mm FWHM 40 mm from the collimator surface, and the sensitivity was {approx}0.05%. Phantom and small animal images were successfully obtained with our developed gamma camera. Conclusions: These results confirmed that the developed pixelated GSO gamma camera has potential as an effective instrument for low energy gamma photon imaging.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Accurate, Fast and Scalable Kernel Ridge Regression on Parallel and Distributed Systems

Kernel Ridge Regression (KRR) is a fundamental method in machine learning. Given an n-by-d data matrix as input, a traditional implementation requires Θ(n2) memory to form an n-by-n kernel matrix and Θ(n3) flops to compute the final model. These time and storage costs prohibit KRR from scaling up to large datasets. For example, even on a relatively small dataset (a 520k-by-90 input requiring 357 MB), KRR requires 2 TB memory just to store the kernel matrix. Additionally, the reason is that n usually is much larger than d for real-world applications. On the other hand, weak scaling becomes a problem: if we keep d and n/p fixed as p grows (p is # machines), the memory needed grows as Θ(p) per processor and the flops as Θ(p2) per processor. In the perfect weak scaling situation, both the memory needed and the flops grow as Θ(1) per processor (i.e. memory and flops are constant). The traditional Distributed KRR implementation (DKRR) only achieved 0.32% weak scaling efficiency from 96 to 1536 processors. In this work, we propose two new methods to address these problems: the Balanced KRR (BKRR) and K-means KRR (KKRR). These methods consider alternative ways to partition the input dataset into p different parts, generating p different models, and then selecting the best model among them. Compared to a conventional implementation, KKRR2 (optimized version of KKRR) improves the weak scaling efficiency from 0.32% to 38% and achieves a 591x speedup for getting the same accuracy by using the same data and the same hardware (1536 processors). BKRR2 (optimized version of BKRR) achieves a higher accuracy than the current fastest method using less training time for a variety of datasets. For the applications requiring only approximate solutions, BKRR2 improves the weak scaling efficiency to 92% and achieves 3505x speedup (theoretical speedup: 4096x).

97 MATHEMATICS AND COMPUTING↗

MrHyDE v.1.0

SAND2024-01324O MrHyDE, which stands for Multi-resolution Hybridized Differential Equations, is a general-purpose C++ package for the solution of coupled multiphysics and multiscale systems on massively parallel computing systems. MrHyDE is designed to enable moving beyond forward simulation for multiscale applications which includes optimization, control, uncertainty quantification, and stochastic inversion. The framework provides interfaces to several packages within the Trilinos framework and leverages automatic differentiation to enable adjoint capabilities for large-scale, gradient-based optimization. MrHyDE provides automated multiscale capabilities through a subgrid model interface and multiscale Dirichlet-to-Neumann maps. For extreme-scale applications, MrHyDE provides in situ data-compression algorithms to reduce memory requirements while maintaining performance. MrHyDE is a general-purpose, computational framework for the solution of multiscale and multiphysics applications. It uses a combination of structure-preserving, physics-compatible discretizations, fully implicit methods, multi-resolution schemes, or fully explicit methods. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

SciDAC↗

Comparison of leading parallel NAS file systems on commodity hardware

High performance computing has experienced tremendous gains in system performance over the past 20 years. Unfortunately other system capabilities, such as file I/O, have not grown commensurately. In this activity, we present the results of our tests of two leading file systems (GPFS and Lustre) on the same physical hardware. This hardware is the standard commodity storage solution in use at LLNL and, while much smaller in size, is intended to enable us to learn about differences between the two systems in terms of performance, ease of use and resilience. This work represents the first hardware consistent study of the two leading file systems that the authors are aware of.

97 MATHEMATICS AND COMPUTING↗

Massively parallel hierarchical control system and method

An electronic control system is disclosed for controlling individually controllable elements of an external component. In one embodiment the system may include a state translator subsystem for receiving a state command from an external subsystem. The state translator subsystem may have at least one module for processing the state command and generating operational commands for controlling the elements to achieve a desired state or condition. A programmable calibration command translation layer (PCCTL) subsystem may be included which receives and uses the operational commands to generate granular level commands for controlling the elements. A feedback control layer subsystem may be included which applies the granular level commands to the elements, and further modifies the granular level commands as needed to control the elements in closed loop fashion.

Panas, Robert Matthew↗

Valved nozzle with a compensator and massively parallel 3D printing system

In one aspect, the present disclosure provides a nozzle for a 3D printing system. The nozzle may include a flowpath with a material inlet and a material outlet. The nozzle may further include a valve in fluid communication with the flowpath between the material inlet and the material outlet, where the valve includes a closed state and an open state, where in the closed state the valve obstructs the flowpath between the material inlet and the material outlet, and where in the open state the material inlet is in fluid communication with the material outlet. The nozzle may further include a compensator in fluid communication with the flowpath, where the compensator includes a contracted state associated with the open state of the valve and an expanded state associated with the closed state of the valve.

Lewis, Jennifer A.↗

Detailed Modeling, Design, and Evaluation of a Scalable Multi-level Checkpointing System

High-performance computing (HPC) systems are growing more powerful by utilizing more hardware components. As the system mean-time-before-failure correspondingly drops, applications must checkpoint more frequently to make progress. However, as the system memory sizes grow faster than the bandwidth to the parallel file system, the cost of checkpointing begins to dominate application run times. A potential solution to this problem is to use multi-level checkpointing, which employs multiple types of checkpoints with different costs and different levels of resiliency in a single run. The goal is to design light-weight checkpoints to handle the most common failure modes and rely on more expensive checkpoints for less common, but more severe failures. While this approach is theoretically promising, it has not been fully evaluated in a large-scale, production system context. To this end we have designed a system, called the Scalable Checkpoint/Restart (SCR) library, that writes checkpoints to storage on the compute nodes utilizing RAM, Flash, or disk, in addition to the parallel file system. We present the performance and reliability properties of SCR as well as a probabilistic Markov model that predicts its performance on current and future systems. We show that multi-level checkpointing improves efficiency on existing large-scale systems and that this benefit increases as the system size grows. In particular, we developed low-cost checkpoint schemes that are 100x-1000x faster than the parallel file system and effective against 85% of our system failures. This leads to a gain in machine efficiency of up to 35%, and it reduces the the load on the parallel file system by a factor of two on current and future systems.

99 GENERAL AND MISCELLANEOUS↗

IOrchestrator: improving the performance of multi-node I/O systems via inter-server coordination

A cluster of I/O nodes and a parallel file system are often used to provide high-throughput I/O service to a parallel compute cluster. To exploit I/O parallelism parallel file systems stripe file data across the I/O nodes. While this practice is effective in serving asynchronous requests, it may break individual program's spatial locality, which can seriously degrade I/O performance when the I/O nodes concurrently serve synchronous requests from multiple I/O-intensive programs. In this paper we propose a scheme, Orchestrator, to improve the I/O performance of multi-I/O-node systems by orchestrating I/O services among programs when such inter-I/O-node coordination is dynamically determined to be cost effective. We have implemented IOrchestrator in the PVFS2 parallel file system. Our experiments with representative parallel benchmarks show that IOrchestrator can significantly improve I/O performance - by up to a factor of 2.5 - delivered by a cluster of I/O nodes servicing concurrently-running parallel programs. Notably, we have not observed any scenarios in which the use of IOrchestrator causes significant performance degradation.

97 MATHEMATICS AND COMPUTING↗

Layout-Aware I/O Scheduling for Terabits Data Movement

Many science facilities, such as the Department of Energy s Leadership Computing Facilities and experimental facilities including the Spallation Neutron Source, Stanford Linear Accelerator Center, and Advanced Photon Source, produce massive amounts of experimental and simulation data. These data are often shared among the facilities and with collaborating institutions. Moving large datasets over the wide- area network (WAN) is a major problem inhibiting collaboration. Next- generation, terabit-networks will help alleviate the problem, however, the parallel storage systems on the end-system hosts at these institutions can become a bottleneck for terabit data movement. The parallel storage system (PFS) is shared by simulation systems, experimental systems, analysis and visualization clusters, in addition to wide-area data movers. These competing uses often induce temporary, but significant, I/O load imbalances on the storage system, which impact the performance of all the users. The problem is a serious concern because some resources are more expensive (e.g. super computers) or have time-critical deadlines (e.g. experimental data from a light source), but parallel file systems handle all requests fairly even if some storage servers are under heavy load. This paper investigates the problem of competing workloads accessing the parallel file system and how the performance of wide-area data movement can be improved in these environments. First, we study the I/O load imbalance problems using actual I/O performance data collected from the Spider storage system at the Oak Ridge Leadership Computing Facility. Second, we present I/O optimization solutions with layout-awareness on end-system hosts for bulk data movement. With our evaluation, we show that our I/O optimization techniques can avoid the I/O congested disk groups, improving storage I/O times on parallel storage systems for terabit data movement.

I/O and Storage↗