Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “I/O performance”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Efficient Asynchronous I/O with Request Merging

With the advancement of exascale computing, the amount of scientific data is increasing day by day. Efficient data access is necessary for scientific discoveries. Unfortunately, the I/O performance is not improved, like the CPU and network speed. So, I/O operations take longer time than data generation or analysis. Asynchronous I/O has been proposed to extenuate the I/O bottleneck by overlapping I/O and computation time. However, multiple small write operations can diminish the benefits of asynchronous I/O, as the I/O time becomes significantly longer than the compute time, with little time to overlap with. To overcome these issues, we present an optimization technique to merge small contiguous write operations. We integrated our solution into the HDF5 asynchronous I/O VOL connector and demonstrated the effectiveness of merging HDF5 write operations automatically and transparently without requiring any code change from the application.

Chowdhury, Kamal Hossain↗
NASA NTRS1995

Performance Modeling of Network-Attached Storage Device Based Hierarchical Mass Storage Systems

Network attached storage devices improve I/O performance by separating control and data paths and eliminating host intervention during the data transfer phase. Devices are attached to both a high speed network for data transfer and to a slower network for control messages. Hierarchical mass storage systems use disks to cache the most recently used files and a combination of robotic and manually mounted tapes to store the bulk of the files in the file system. This paper shows how queuing network models can be used to assess the performance of hierarchical mass storage systems that use network attached storage devices as opposed to host attached storage devices. Simulation was used to validate the model. The analytic model presented here can be used, among other things, to evaluate the protocols involved in 1/0 over network attached devices.

Menasce, Daniel A.↗
DOE OSTI2022

Performance Analysis and Optimization for Scientific Data Workloads

Scientific data generated at experimental and observational facilities are increasingly being processed on large-scale compute systems. Most of the experimental data analysis workflows are not designed or implemented to run on large scale environments and take full advantage of HPC compute and storage resources. These applications are unlike the traditional tightly-coupled scientific applications and hence face significant performance and scalability challenges as the volume of data increases exponentially. In this paper, we conduct a performance and scalability analysis for experimental analysis applications and workflows operating on data from light sources. Our analysis detects and quantifies I/O performance, scalability and runtime bottlenecks for three data analysis applications that run on NERSC resources. Based on our analysis we propose and implement a set of optimizations that lead to reducing the amount of time spent on I/O operations by almost 90%.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

OCTOKV: An Agile Network-Based Key-Value Storage System with Robust Load Orchestration

In this paper, we propose OctoKV, an innovative network-based key-value storage system. OctoKV addresses the repetitive address translation overhead associated with traditional key-value stores running on file systems on the client side. To mitigate this overhead, we implemented the key-value store on the server side using NVMe-oF and a user-level NVMe driver. In particular, we employed fine-grained resource monitoring and load balancing based on heuristics to optimize I/O performance. OctoKV is deployed on a Linux cluster with Intel SPDK. The extensive evaluation shows that OctoKV achieves lower I/O response times in comparison to traditional approaches where key-value stores run on the client side. Also, the proposed load balancing strategies efficiently enhance I/O response times by equally distributing the workload from overloaded cores to other cores.

Khan, Awais↗
NASA NTRS2010

Application-Controlled Parallel Asynchronous Input/Output Utility

A software utility tool has been designed to alleviate file system I/O performance bottlenecks to which many high-end computing (HEC) applications fall prey because of the relatively large volume of data generated for a given amount of computational work. In an effort to reduce computing resource waste, and to improve sustained performance of these HEC applications, a lightweight software utility has been designed to circumvent bandwidth limitations of typical HEC file systems by exploiting the faster inter-processor bandwidth to move output data from compute nodes to designated I/O nodes as quickly as possible, thereby minimizing the I/O wait time. This utility has successfully demonstrated a significant performance improvement within a major NASA weather application.

Clune, Thomas↗
DOE OSTI2020

GPU Direct I/O with HDF5

Exascale HPC systems are being designed with accelerators, such as GPUs, to accelerate parts of applications. In machine learning workloads as well as large-scale simulations that use GPUs as accelerators, the CPU (or host) memory is currently used as a buffer for data transfers between GPU (or device) memory and the file system. If the CPU does not need to operate on the data, then this is sub-optimal because it wastes host memory by reserving space for duplicated data. Furthermore, this “bounce buffer” approach wastes CPU cycles spent on transferring data. A new technique, NVIDIA GPUDirect Storage (GDS), can eliminate the need to use the host memory as a bounce buffer. Thereby, it becomes possible to transfer data directly between the device memory and the file system. This direct data path shortens latency by omitting the extra copy and enables higher-bandwidth. To take full advantage of GDS in existing applications, it is necessary to provide support with existing I/O libraries, such as HDF5 and MPI-IO, which are heavily used in applications. In this paper, we describe our effort of integrating GDS with HDF5, the top I/O library at NERSC and at DOE leadership computing facilities. We design and implement this integration using a HDF5 Virtual File Driver (VFD). The GDS VFD provides a file system abstraction to the application that allows HDF5 applications to perform I/O without the need to move data between CPUs and GPUs explicitly. We compare performance of the HDF5 GDS VFD with explicit data movement approaches and demonstrate superior performance with the GDS method.

Ravi, J↗
DOE OSTI2022

Hvac: Removing I/O Bottleneck for Large-Scale Deep Learning Applications

Scientific communities are increasingly adopting deep learning (DL) models in their applications to accelerate scientific discovery processes. However, with rapid growth in the computing capabilities of HPC supercomputers, large-scale DL applications have to spend a significant portion of training time performing I/O to a parallel storage system. Previous research works have investigated optimization techniques such as prefetching and caching. Unfortunately, there exist non-trivial challenges to adopting the existing solutions on HPC supercomputers for large-scale DL training applications, which include non-performance and/or failures at extreme scale, lack of portability and generality in design, complex deployment methodology, and being limited to a specific application or dataset. To address these challenges, we propose High-Velocity AI Cache (HVAC), a distributed read-cache layer that targets and fully exploits the node-local storage or near node-local storage technology. HVAC seamlessly accelerates read I/O by aggregating node-local or near node-local storage, avoiding metadata lookups and file locking while preserving portability in the application code. We deploy and evaluate HVAC on 1,024 nodes (with over 6000 NVIDIA V100 GPUS) of the Summit supercomputer. In particular, we evaluate the scalability, efficiency, accuracy, and load distribution of HVAC compared to GPFS and XFS-on-NVMe. With four different DL applications, we observe an average 25 % performance improvement atop GPFS and 9% drop against XFS-on-NVMe, which scale linearly and are considered the performance upper bound. We envision HVAC as an important caching library for upcoming HPC supercomputers such as Frontier.

Khan, Awais↗
DOE OSTI2026

Scientific Data Compression for Large Scale Computational Fluid Dynamics (CFD) Simulations

This Cooperative Research and Development Agreement (CRADA) between Oak Ridge National Laboratory (ORNL) and General Electric (GE) investigated methods for reducing the size of large computational fluid dynamics (CFD) simulation datasets using scientific data compression techniques. The work focused on adapting the MultiGrid Adaptive Reduction of Data (MGARD) compression framework and integrating it with high-performance I/O and visualization tools used in CFD workflows. MGARD uses hierarchical multilevel decomposition to enable error-controlled compression of floating-point scientific data while preserving quantities of interest. During the project, MGARD compression was integrated with the ADIOS I/O framework and visualization tools such as ParaView to enable efficient storage, transfer, and analysis of simulation data. The collaboration also explored approaches for improving compression performance for CFD data defined on unstructured meshes. Results demonstrate that scientific data compression can significantly reduce storage requirements and improve data management for large-scale CFD simulations.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1996

RANS-MP: A Portable Parallel Navier-Stokes Solver

RANS-MP, a new implementation of a single-grid Navier-Stokes solver using the diagonalized Beam-Warming approximate-factorization scheme, is presented. This first release of the completely rewritten solver employs the following optimizations: (1) Bi-directional multi-partition method for the ADI solver part; this improves granularity and load balance; (2) Improved cache usage through elimination of non-unit-stride array access (possible in part due to multi-partitioning); (3) Preprocessing of communicating boundary conditions to streamline logic during time stepping; (4) Truly parallel, high-performance I/O using the newly-developed MPI-IO library; (5) Elimination of large amounts of redundant operations through efficient use of workspace. Results of some realistic wing computations on the IBM SP2 computer will be presented. We will demonstrate that excellent absolute performance and scalability are obtained with RANS-MP, even for relatively small grid sizes. Besides high performance, an outstanding feature of RANS-MP is its true portability, due to the use of the portable message passing and I/O libraries MPI and MPI-IO.

VanderWijngaart, Rob F.↗
DOE OSTI2025

Optimizing Management of Persistent Data Structures in High-Performance Analytics

Large-scale data analytics workflows ingest massive input data into various data structures, including graphs and key-value datastores. These data structures undergo multiple transformations and computations and are typically reused in incremental and iterative analytics workflows. Persisting in-memory views of these data structures enables reusing them beyond the scope of a single program run while avoiding repetitive raw data ingestion overheads. Memory-mapped I/O enables persisting in-memory data structures without data serialization and deserialization overheads. However, memory-mapped I/O lacks the key feature of persisting consistent snapshots of these data structures for incremental ingestion and processing. The obstacles to efficient virtual memory snapshots using memory-mapped I/O include background writebacks outside the application’s control, and the significantly high storage footprint of such snapshots. To address these limitations, we present Privateer, a memory and storage management tool that enables storage-efficient virtual memory snapshotting while also optimizing snapshot I/O performance. Here, we integrated Privateer into Metall, a state-of-the-art persistent memory allocator for C++, and the Lightning Memory-Mapped Database (LMDB), a widely-used key-value datastore in data analytics and machine learning. Privateer optimized application performance by 1.22× when storing data structure snapshots to node-local storage, and up to 16.7× when storing snapshots to a parallel file system. Privateer also optimizes storage efficiency of incremental data structure snapshots by up to 11× using data deduplication and compression.

Computer science↗
DOE OSTI2022

Organizing Large Data Sets for Efficient Analyses on HPC Systems

Upcoming exascale applications could introduce significant data management challenges due to their large sizes, dynamic work distribution, and involvement of accelerators such as graphical processing units, GPUs. In this work, we explore the performance of reading and writing operations involving one such scientific application on two different supercomputers. Our tests showed that the Adaptable Input and Output System, ADIOS, was able to achieve speeds over 1TB/s, a significant fraction of the peak I/O performance on Summit. We also demonstrated the querying functionality in ADIOS could effectively support common selective data analysis operations, such as conditional histograms. In tests, this query mechanism was able to reduce the execution time by a factor of five. More importantly, ADIOS data management framework allows us to achieve these performance improvements with only a minimal amount of coding effort.

Gu, Junmin↗
DOE OSTI2022

CEAZ: Accelerating Parallel I/O Via Hardware-Algorithm Co-Designed Adaptive Lossy Compression

As supercomputers continue to grow to exa-scale, the amount of data that needs to be saved or transmitted is exploding. To this end, many previous works have studied using error-bounded lossy compressors to reduce the data size and improve the I/O performance. However, little work has been done for effectively offloading lossy compression onto FPGA-based SmartNICs to reduce the compression overhead. In this paper, we propose a hardware-algorithm co-design of efficient and adaptive lossy compressor for scientific data on FPGAs (called CEAZ) to accelerate parallel I/O. Our contribution is fourfold: (1) We propose an efficient Huffman coding approach that can adaptively update Huffman codewords online based on codewords generated offline (from a variety of representative scientific datasets). (2) We derive a theoretical analysis to support a precise control of compression ratio under an error-bounded compression mode, enabling accurate offline Huffman codewords generation. This also help us create a fixed-ratio compression mode for consistent throughput. (3) We develop an efficient compression pipeline by adopting cuSZ’s dual-quantization algorithm to our hardware use case. (4) We evaluate CEAC on five real-world datasets with both a single FPGA board and 256 nodes from Bridges2 supercomputer. Experiments show that CEAZ outperforms the second-best FPGA-based lossy compressor by 2× of throughput and 9.6× of compression ratio. It also improves MPI_File_write and MPI_Gather throughputs by up to 32.7× and 31.4×, respectively.

Zhang, Chengming↗
DOE OSTI2024

Accelerating Application Bulk Synchronous Writes in HPC Environments

High-bandwidth storage tiers are becoming more common for their capability to absorb high-rate, bursty I/Os. Notably, the designs of these fast storage tiers differ from system to system. The variation of these layers and non-uniform methods of access can pose chal- lenges for applications seeking to run at multiple HPC facilities. Therefore, in this work, we present Spectral, a rapid-output ab- straction library to accelerate application, bulk-synchronous writes on HPC systems. We design Spectral to enable applications to use high-bandwidth storage, such as node-local storage and dis- tributed, write-caches (e.g., burst buffers) transparently without requiring modifications to the application or file system source code. The key idea is to allow applications to spend most of the time performing productive work and to not require any source code changes for maximum portability on different HPC archi- tectures. Spectral internally re-routes write-only files through available, high-performance I/O resources before ultimately mi- grating them to the shared global parallel file system. For instance, on Summit, Spectral transparently places application outputs on node-local storage and then utilizes asynchronous migration to the center-wide GPFS file system. We evaluate Spectral on the Summit HPC system (1024 nodes) using the IOR benchmark and real scientific applications. Spectral shows linear performance scaling, improving application write performance by over an order of magnitude when compared to GPFS.

Khan, Awais↗
DOE OSTI2020

High Performance Computing Application I/O Traces

The dataset comprises trace files from high performance computing (HPC) simulations. The trace files contain records of every I/O operation executed by a simulation application run, including I/O operations from HDF5, MPI-IO, and POSIX and all of the parameters supplied to those operations, e.g. file name, offset, and flags. The traces are generated by executing a simulation application that is linked with the Recorder tracing tool (https://github.com/uiuc-hpc/Recorder). The Recorder trace tool intercepts the I/O calls made by the application, records the I/O trace record, and then calls the intended I/O call so that the operation executes.

Wang, C.↗
NASA NTRS1993

Evaluation Metrics for the Paragon XP/S-15

On February 17th 1993, the Numerical Aerodynamic Simulation (NAS) facility located at the NASA Ames Research Center installed a 224 node Intel Paragon XP/S-15 system. After its installation, the Paragon was found to be in a very immature state and was unable to support a NAS users' workload, composed of a wide range of development and production activities. As a first step towards addressing this problem, we implemented a set of metrics to objectively monitor the system as operating system and hardware upgrades were installed. The metrics were designed to measure four aspects of the system that we consider essential to support our workload: availability, utilization, functionality, and performance. This report presents the metrics collected from February 1993 to August 1993. Since its installation, the Paragon availability has improved from a low of 15% uptime to a high of 80%, while its utilization has remained low. Functionality and performance have improved from merely running one of the NAS Parallel Benchmarks to running all of them faster (between 1 and 2 times) than on the iPSC/860. In spite of the progress accomplished, fundamental limitations of the Paragon operating system are restricting the Paragon from supporting the NAS workload. The maximum operating system message passing (NORMA IPC) bandwidth was measured at 11 Mbytes/s, well below the peak hardware bandwidth (175 Mbytes/s), limiting overall virtual memory and Unix services (i.e. Disk and HiPPI I/O) performance. The high NX application message passing latency (184 microns), three times than on the iPSC/860, was found to significantly degrade performance of applications relying on small message sizes. The amount of memory available for an application was found to be approximately 10 Mbytes per node, indicating that the OS is taking more space than anticipated (6 Mbytes per node).

Traversat, Bernard↗
NASA NTRSDate not supplied

NASA POWER: Providing Analysis-Ready, Cloud-Optimized Data for AI /ML Training and Applications in Earth Science

As global demand for sustainable development grows, the integration of Earth Observation (EO) data into decision making frameworks has become a primary objective for the scientific community. The NASA Prediction of Worldwide Energy Resources (POWER) project serves as a bridge between NASA EO data and the specialized needs of the renewable energy, sustainable infrastructure and agroclimatology communities. In this poster presentation we will present an overview of POWER data products and services along with its use in diverse research to decision-making workflows. By providing over 40 years of high-resolution historical, hourly and daily solar and meteorological data, POWER transforms satellite observations and global model reanalysis into actionable, Analysis-Ready Dataset (ARD). Currently, the project delivers over 250 industry-friendly parameters to the users from different NASA datasets like CERES SYN1Deg, MERRA-2, and IMERG alongside downscaled CMIP6 climate model data, fulfilling over 16 million requests from 50,000 unique users monthly. To ensure data quality and traceability, these parameters are rigorously validated against the ground-based observations from the Baseline Surface Radiation Network (BSRN) and the Global Surface Summary of the Day (GSOD) – these results will be discussed in the presentation. A newly introduced web-based PaRameter Uncertainty ViEwer (PRUVE) tool will be presented that provides an online validation platform to the users that benchmarks satellite-based and assimilation data products against these surface measurements. To reduce technical barriers to data adoption, POWER data is accessible through RESTful APIs, ESRI ArcGIS Image Services, a web-based Data Access Viewer tool, allowing users to visualize, validate and apply the dataset. For efficient data delivery POWER data is cloud-optimized into Zarr datastore accessible through NASA managed Amazon S3 ensures high-performance allowing users to integrate EO directly into operational pipelines. These customized services will be presented. Use cases from application will be presented from the energy sector - such as for design of generation systems, performance monitoring of solar power plants, in infrastructure sector- optimizing building energy efficiency and thermal comfort, in agriculture – such as driving crop simulation and yield forecasting models to enable climate resilient farming. Furthermore, the shift toward machine learning (ML) in EO research that has positioned POWER as a key provider for training datasets which will be discussed. Use-cases will be presented to showcase how NASA data is enabling the development of predictive tools for climate variability and resource management. The poster will present POWER’s future plans including technology development to enhance data traceability and reproducibility and improving I/O performance to support the rapid integration of new EO products, ensuring that POWER remains a robust scalable backend for the evolving landscape of AI-driven Earth Science. Additionally, POWER is developing an AI Agent and an MCP-Server to enable industry AI-Agentic workflows.

Neha Khadka↗
NASA NTRS1992

Crystal ball gazing

Over the last seven years, the CPU on my desk has increased speed by two orders of magnitude, from around 1 MIP to more than 100 MIPS; more important is that it is about as fast as any uniprocessor of any type available for any price, for compute bound problems. Memory on the system is also about 100 times as big, while disk is only about 10 times as big. Local network and I/O performance have increased greatly, though not quite at the same rate as processor speed. More important, I will argue, is that the CPU's address space is 64 bits, rather than 32 bits, allowing us to rethink some time honored presumptions. The Internet has gone from a few hundred machines to a million, and now have grown to span the entire globe, and wide area networks have now becoming commercial services. 'PC's' are now real computers, bringing what was top of the line computing capability to the masses only a few years behind the leading edge. So even a year or two from now, we can anticipate commonplace desktop machines running at speeds hundreds of MIPS, with main memories in the hundreds of megabytes to a gigabyte, able to draw millions of vectors/second, and all capable of some reasonable 3D graphics. And only a few years later, this will be the $1500 PC. So the 1990's certainly brings: 64 bit processors becoming standard; BIP/BFLOP class uniprocessors; large scale multiprocessors for special purpose applications; I/O as the most significant computer engineering problem; Hierarchical data servers in everyday use; routine access to archived data around the world; and what else? What do systems such as those we will have this decade imply to those building data analysis systems today? Many of the presumptions of the 1970's and 1980's need to be reexamined in the light of 1990's technology.

Gettys, Jim↗