SEARCH · Search NASA
Results for “Data compression”
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.
CRUSH: The NSI data compression utility
CRUSH is a data compression utility that provides the user with several lossless compression techniques available in a single application. It is intended that the future development of CRUSH will depend upon feedback from the user community to identify new features and capabilities desired by the users. CRUSH provides an extension to the UNIX Compress program and the various VMS implementations of Compress that many users are familiar with. An important capability added by CRUSH is the addition of additional compression techniques and the option of automatically determining the best technique for a given data file. The CRUSH software is written in C and is designed to run on both VMS and UNIX systems. VMS files that are compressed will regain their full file characteristics upon decompression. To the extent possible, compressed files can be transferred between VMS and UNIX systems, and thus be decompressed on a different system than they were compressed on. Version 1 of CRUSH is currently available. This version is a VAX VMS implementation. Version 2, which has the full range of capabilities for both VMS and UNIX implementations, will be available shortly.
Online and Scalable Data Compression Pipeline with Guarantees on Quantities of Interest
Data compression is becoming critical for data-intensive scientific applications. Scientists require compression techniques that accurately preserve derived quantities of interest (QoIs). Prior work has shown that a pipeline can be built to guarantee error on the primary data (PD) within user-defined bounds and achieve near-floating point QoI errors. In this paper, we present novel computational approaches for accelerating the pipeline and demonstrate results that enable concurrent execution of compression in parallel with the simulation nodes. This allows compression, including the writing of the required compression data, for the previous time step to be completed while the simulation proceeds with the current time step. Overall, the approach presented in this paper results in a 6–8 times improvement in computational overhead compared to previous work. These results were obtained using data generated by a large-scale fusion code called XGC, which produces hundreds of terabytes of data in a single day.
The Space and Earth Science Data Compression Workshop
This document is the proceedings from a Space and Earth Science Data Compression Workshop, which was held on March 27, 1992, at the Snowbird Conference Center in Snowbird, Utah. This workshop was held in conjunction with the 1992 Data Compression Conference (DCC '92), which was held at the same location, March 24-26, 1992. The workshop explored opportunities for data compression to enhance the collection and analysis of space and Earth science data. The workshop consisted of eleven papers presented in four sessions. These papers describe research that is integrated into, or has the potential of being integrated into, a particular space and/or Earth science data information system. Presenters were encouraged to take into account the scientists's data requirements, and the constraints imposed by the data collection, transmission, distribution, and archival system.
Subband Coding Methods for Seismic Data Compression
This paper presents a study of seismic data compression techniques and a compression algorithm based on subband coding. The compression technique described could be used as a progressive transmission system, where successive refinements of the data can be requested by the user. This allows seismologists to first examine a coarse version of waveforms with minimal usage of the channel and then decide where refinements are required. Rate-distortion performance results are presented and comparisons are made with two block transform methods.
Efficient I/O for Neural Network Training with Compressed Data
FanStore is a shared object store that enables efficient and scalable neural network training on supercomputers. By providing a global cache layer on node-local burst buffers using a compressed representation, it significantly enhances the processing capability of deep learning (DL) applications on existing hardware. In addition, FanStore allows POSIX-compliant file access to the compressed data in user space. We investigate the tradeoff between runtime overhead and data compression ratio using real-world datasets and applications, and propose a compressor selection algorithm to maximize storage capacity given performance constraints. We consider both asynchronous (i.e., with prefetching) and synchronous I/O strategies, and propose mechanisms for selecting compressors for both approaches. Using FanStore, the same storage hardware can host 2-13x more data for example applications without significant runtime overhead. Empirically, our experiments show that FanStore scales to 512 compute nodes with near linear performance scalability.
Efficient Data Compression for 3D Sparse TPC via Bicephalous Convolutional Autoencoder
Real-time data collection and analysis in large experimental facilities present a great challenge across multiple domains, including high energy physics, nuclear physics, and cosmology. To address this, machine learning (ML)-based methods for real-time data compression have drawn significant attention. However, unlike natural image data, such as CIFAR and ImageNet that are relatively small-sized and continuous, scientific data often come in as three-dimensional 3D data volumes at high rates with high sparsity (many zeros) and non-Gaussian value distribution. This makes direct application of popular ML compression methods, as well as conventional data compression methods, suboptimal. To address these obstacles, this work introduces a dual-head autoencoder to resolve sparsity and regression simultaneously, called Bicephalous Convolutional AutoEncoder (BCAE). This method shows advantages both in compression fidelity and ratio compared to traditional data compression methods, such as MGARD, SZ, and ZFP. To achieve similar fidelity, the best performer among the traditional methods can reach only half the compression ratio of BCAE. Moreover, a thorough ablation study of the BCAE method shows that a dedicated segmentation decoder improves the reconstruction.
Adaptive data compression for video signals
Adaptive data compression for video signals using Tiros television cloud pictures
A Lookup Table Based Low Resource Usage Waveform Data Compression Block in FPGA
aA lookup table based lossless data compression scheme suitable for FPGA implementation, with very low silicon resource usage, is developed. Different algorithms can be implemented by simply changing the contents of the lookup tables in the functional block while keeping the logic circuits unchanged. It exhibits performance comparable to the zip compression software in Linux distributions. The compression block can fit into almost any existing FPGA used in the online data acquisition systems of high energy physics experiments. The decompression process can be performed in either software or in FPGA firmware, allowing users to process compressed data flexibly.
Data compression for the Cassini radio and plasma wave instrument
The Cassini Radio and Plasma Wave Science experiment will employ data compression to make effective use of the available data telemetry bandwidth. Some compression will be achieved by use of a lossless data compression chip and some by software in a dedicated 80C85 processor. A description of the instrument and data compression system are included in this report. Also, the selection of data compression systems and acceptability of data degradation is addressed.
The implementation of a lossless data compression module in an advanced orbiting system: Analysis and development
Data compression has been proposed for several flight missions as a means of either reducing on board mass data storage, increasing science data return through a bandwidth constrained channel, reducing TDRSS access time, or easing ground archival mass storage requirement. Several issues arise with the implementation of this technology. These include the requirement of a clean channel, onboard smoothing buffer, onboard processing hardware and on the algorithm itself, the adaptability to scene changes and maybe even versatility to the various mission types. This paper gives an overview of an ongoing effort being performed at Goddard Space Flight Center for implementing a lossless data compression scheme for space flight. We will provide analysis results on several data systems issues, the performance of the selected lossless compression scheme, the status of the hardware processor and current development plan.
Low-Complexity Lossless and Near-Lossless Data Compression Technique for Multispectral Imagery
This work extends the lossless data compression technique described in Fast Lossless Compression of Multispectral- Image Data, (NPO-42517) NASA Tech Briefs, Vol. 30, No. 8 (August 2006), page 26. The original technique was extended to include a near-lossless compression option, allowing substantially smaller compressed file sizes when a small amount of distortion can be tolerated. Near-lossless compression is obtained by including a quantization step prior to encoding of prediction residuals. The original technique uses lossless predictive compression and is designed for use on multispectral imagery. A lossless predictive data compression algorithm compresses a digitized signal one sample at a time as follows: First, a sample value is predicted from previously encoded samples. The difference between the actual sample value and the prediction is called the prediction residual. The prediction residual is encoded into the compressed file. The decompressor can form the same predicted sample and can decode the prediction residual from the compressed file, and so can reconstruct the original sample. A lossless predictive compression algorithm can generally be converted to a near-lossless compression algorithm by quantizing the prediction residuals prior to encoding them. In this case, since the reconstructed sample values will not be identical to the original sample values, the encoder must determine the values that will be reconstructed and use these values for predicting later sample values. The technique described here uses this method, starting with the original technique, to allow near-lossless compression. The extension to allow near-lossless compression adds the ability to achieve much more compression when small amounts of distortion are tolerable, while retaining the low complexity and good overall compression effectiveness of the original algorithm.
Device and method for data compression using a metadata cache
A processing device is provided which includes memory comprising data cache memory configured to store compressed data and metadata cache memory configured to store metadata, each portion of metadata comprising an encoding used to compress a portion of data. The processing device also includes at least one processor configured to compress portions of data and select, based on one or more utility level metrics, portions of metadata to be stored in the metadata cache memory. The at least one processor is also configured to store, in the metadata cache memory, the portions of metadata selected to be stored in the metadata cache memory, store, in the data cache memory, each portion of compressed data having a selected portion of corresponding metadata stored in the metadata cache memory. Each portion of compressed data, having the selected portion of corresponding metadata stored in the metadata cache memory, is decompressed.
Channel coding and data compression system considerations for efficient communication of planetary imaging data
End-to-end system considerations involving channel coding and data compression which could drastically improve the efficiency in communicating pictorial information from future planetary spacecraft are presented.
Channel coding and data compression system considerations for efficient communication of planetary imaging data
End-to-end system considerations involving channel coding and data compression are reported which could drastically improve the efficiency in communicating pictorial information from future planetary spacecraft. In addition to presenting new and potentially significant system considerations, this report attempts to fill a need for a comprehensive tutorial which makes much of this very subject accessible to readers whose disciplines lie outside of communication theory.
Proceedings of the Scientific Data Compression Workshop
Continuing advances in space and Earth science requires increasing amounts of data to be gathered from spaceborne sensors. NASA expects to launch sensors during the next two decades which will be capable of producing an aggregate of 1500 Megabits per second if operated simultaneously. Such high data rates cause stresses in all aspects of end-to-end data systems. Technologies and techniques are needed to relieve such stresses. Potential solutions to the massive data rate problems are: data editing, greater transmission bandwidths, higher density and faster media, and data compression. Through four subpanels on Science Payload Operations, Multispectral Imaging, Microwave Remote Sensing and Science Data Management, recommendations were made for research in data compression and scientific data applications to space platforms.
Advances in data compression
A rigorous real-variables treatment of general data compression and encoding problems, centered on formulation and proof of relevant existence theorems and a unified formulation of source coding (both noiseless and with a fidelity criterion) in inaccurately or incompletely specified statistical environments. Difficulties in modeling of sources with unknown or imperfectly known statistical descriptions are analyzed and source codes (SC) are classified (variable-rate noiseless SC, fixed-rate noiseless SC) and analyzed, along with types of code sequences (weighted-universal, maximin-universal, strongly or weakly minimax-universal). Universal coding on video data, variable-rate coding with distortion, and distortion-rate functions are discussed. Design strategies for universal coding algorithms are suggested, but the article is not oriented to specific methods of synthesizing data compression systems.
In-pixel AI for lossy data compression at source for X-ray detectors
Integrating neural networks for data compression directly in the Read-Out Integrated Circuits (ROICs), i.e. the pixelated front-end, would result in a significant reduction in off-chip data transfer, overcoming the I/O bottleneck. Our ROIC test chip (AI-In-Pixel-65) is designed in a 65 nm Low Power CMOS process for the readout of pixelated X-ray detectors. Each pixel consists of an analog front-end for signal processing and a 10b analog-to-digital converter operating at 100KSPS. Here, we compare two non-reconfigurable techniques, Principal Component Analysis (PCA) and an AutoEncoder (AE) as lossy data compression engines implemented within the pixelated area. The PCA algorithm achieves 50$×$ compression, adds one clock cycle latency, and results in a 21% increase in the pixel area. The AE achieves 70$×$ compression, adds 30 clock cycle latency, and results in a similar area increase.