Search NASA⌕ Search

SEARCH · Search NASA

Results for “Writing”

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 1,603 records · Page 89

The SENSEI Generic In Situ Interface, In: ISAV '16 Proceedings of the 2nd Workshop on In Situ Infrastructures for Enabling Extreme-scale Analysis and Visualization

The SENSEI generic in situ interface is an API that promotes code portability and reusability. From the simulation view, a developer can instrument their code with the SENSEI API and then make make use of any number of in situ infrastructures. From the method view, a developer can write an in situ method using the SENSEI API, then expect it to run in any number of in situ infrastructures, or be invoked directly from a simulation code, with little or no modification. This paper presents the design principles underlying the SENSEI generic interface, along with some simplified coding examples.

Computer Science↗

UltraLiM: In-Memory Boolean Logic Architecture Using UltraRAM

Conventional computing architectures encounter ‘von Neumann’ and ‘memory wall’ bottlenecks which arise due to the back-and-forth data movement between the physically separate memory and processing units and the speed mismatch between them, respectively. These bottlenecks hurt both energy efficiency and the throughput of computing systems. To address these challenges, in-memory computing architectures have emerged as a promising alternative. They reduce the need for frequent data movement by executing different computing tasks inside the memory system. Here, we present UltraLiM, a logic-in-memory architecture using the UltraRAM-based memory system. UltraRAM holds the promise of developing a ‘universal memory’, overcoming the limitations of charge-based memories thanks to their non-volatile behavior with lower operating voltage. This work presents an in-memory computing architecture that integrates an UltraRAM-based memory array with a custom-designed peripheral circuitry. With this architecture, we can perform various in-memory Boolean logic operations (such as NOT, NAND, NOR, and XOR) in a single cycle. Leveraging the separate read-write paths in the UltraRAM-based memory array, we optimize read operations without encountering design conflicts. This optimization enhances the sense margin, enabling the use of simpler peripheral circuitry for in-memory logic operations.

Alam, Shamiul [University of Tennessee, Knoxville ↗

Distributed Fiber Sensors With High Spatial Resolution in Extreme Radiation Environments in Nuclear Reactor Cores

This paper is a comprehensive experimental report on the neutron radiation effects of distributed optical fiber sensors with enhanced Rayleigh scattering profiles in an in-pile environment. Femtosecond laser direct writing was used to inscribe Type-II modifications in standard telecom fibers and radiation-hardened fibers with fluorine-doped cores. Rayleigh backscattering signals were enhanced for continuous 1.5 m. In-pile lead-out sensors tests were carried out at the MIT Research Reactor for two months, which was operated at a nominal power of 5.7 MW with fast neutron (>0.1 MeV) flux of 1.29 × 1014 n/cm2/s and an in-core temperature of up to 560°C. Using the Optical Frequency Domain Reflectometry technique, the backscattering profiles of fiber sensors were interrogated with a 3-cm spatial resolution to monitor the temperature profile of the reactor. Results show that laser inscribed Type-II modifications in the form of nanogratings are highly stable against extreme temperature and ionizing radiation. Both standard telecom fibers and radiation-hardened fibers with laser-enhanced Rayleigh profiles can continuously perform distributed temperature measurements over the entire duration of the in-pile testing. Temperature coefficients of sensors and spectral shift quality were studied as functions of total radiation fluence. To the best of our knowledge, we present for the first time, the temperature profile of an operating nuclear reactor core with 3-cm spatial resolution, enabled by distributed fiber sensors with laser-enhanced Rayleigh scattering profiles. The high spatial resolution measurements can provide valuable data for the design and validation of digital twin and virtual reality of nuclear energy systems.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Toward an Electromagnetic Event Resilient Grid

The motivation for writing this article is that the threat of a large-scale electromagnetic event having a negative impact on the electric grid is real. Whether human-caused, via the detonation of a nuclear device, or natural, via a high-intensity burst of solar radiation, our historic experience with these phenomena indicates that as a global community, we should be prepared for such events and know how to mitigate their impacts. Current studies and related discussions provide a wide range of damage assessments for these events. We recommend continuing current technical investigations and research as well as strengthening collaboration between stakeholders and experts. This would ensure future threats are addressed in a timely and effective manner.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Improving OFDR Distributed Fiber Sensing by Fibers With Enhanced Rayleigh Backscattering and Image Processing

This article investigates using optical fibers with enhanced backscattering profiles to improve distributed fiber sensor performance and reduce instrumentation costs. Using a femtosecond (fs) laser direct writing technique, the Rayleigh backscattering profile of a standard telecom fiber was enhanced by more than 40 dB to improve the signal-to-noise ratio (SNR) for optical frequency domain reflectometry (OFDR). The enhanced backscattering signals enable effectively distributed strain measurements using a low-cost tunable laser (TL). Median filtering is applied to denoise cross correlation results to further improve measurement outcomes. Results presented in this article show that a TL with a 1-nm tuning range, which is far less than the tuning range used in commercial OFDR interrogators, can perform effectively distributed strain measurements using sensing fibers with enhanced backscattering profiles. The sensing fiber with over 40-dB backscattering enhancement achieved 4.8-cm spatial resolution in strain measurements with a root mean square accuracy of less than 2.70 με when 10–50 με were exerted to the sensing fiber. In conclusion, results presented in this article reveal both the potential and limitations of sensing fibers with enhanced backscattering for OFDR-based distributed fiber sensors.

42 ENGINEERING↗

Design Space Exploration of Ferroelectric Tunnel Junction Toward Crossbar Memories

We perform a simulation-based analysis on the potential of emerging ferroelectric tunnel junctions (FTJs) as a memory device for crossbar arrays. Though FTJs are promising due to their low power switching characteristics compared to other emerging technologies, the greatest challenge for FTJs is the tradeoff between integration density and read performance. Our analysis highlights the need to co-optimize the ferroelectric thickness of the FTJ and read/write voltages to achieve proper functionality at large array sizes. Our analysis shows that FTJ-based crossbar achieves 93% higher sense margin at isoread power of 116 nW (per bit), but this FTJ design comes at a cost of 9.28× higher write power at isowrite time of 250 ns. In response, we study the potential tradeoffs of design points outside the feasible region to understand what device characteristics are desired to overcome such challenges.

Jao, Nicholas↗

Persistent Memory Object Storage and Indexing for Scientific Computing

This paper presents Mosiqs, a persistent memory object storage framework with metadata indexing and querying for scientific computing. We design Mosiqs based on the key idea that memory objects on shared PM pool can live beyond the application lifetime and can become the sharing currency for applications and scientists. Mosiqs provides an aggregate memory pool atop an array of persistent memory devices to store and access memory objects. Mosiqs uses a lightweight persistent memory key-value store to manage the metadata of memory objects such as persistent pointer mappings, which enables memory object sharing for effective scientific collaborations. Mosiqs is implemented atop PMDK. We evaluate the proposed approach on many-core server with an array of real PM devices. The preliminary evaluation confirms a 100% improvement for write and 30% in read performance against a PM-aware file system approach.

Khan, Awais↗

Automated Fortran–C++ Bindings for Large-Scale Scientific Applications

Although many active scientific codes use modern Fortran, most contemporary scientific software libraries are implemented in C and C++. Providing their numerical, algorithmic, or data management features to Fortran codes requires writing and maintaining substantial amounts of glue code. This paper introduces a tool that automatically generates native Fortran 2003 interfaces to C and C++ libraries. The tool supports C++ features that have no direct Fortran analog, such as templated functions and exceptions. Finally, a set of simple examples demonstrates the utility and scope of the tool, and timing measurements with a mock numerical library illustrate the minimal performance impact of the generated wrapper code.

97 MATHEMATICS AND COMPUTING↗

Transient Simulations with a Large Penetration of Converter-Interfaced Generation: Scientific Computing Challenges and Opportunities

Current trends in energy systems point to renewable energy sources (RESs) and battery energy storage systems (BESSs) becoming prevalent in power system operations. As of writing this article, the United States has more than 37 GW of utility-scale solar capacity and an additional 112 GW under development. With the rapidly declining capital costs of many of these technologies, we can expect significant deployments in the coming years.

analytical models↗

Bridging Python to Silicon: The SODA Toolchain

Systems performing scientific computing, data analysis, and machine learning tasks have a growing demand for application-specific accelerators that can provide high computational performance while meeting strict size and power requirements. However, the algorithms and applications that need to be accelerated are evolving at a rate that is incompatible with manual design processes based on hardware description languages. Agile hardware design tools based on compiler techniques can help by quickly producing an application-specific integrated circuit (ASIC) accelerator starting from a high-level algorithmic description. Here, we present the software-defined accelerator (SODA) synthesizer, a modular and open-source hardware compiler that provides automated end-to-end synthesis from high-level software frameworks to ASIC implementation, relying on multilevel representations to progressively lower and optimize the input code. Our approach does not require the application developer to write any register-transfer level code, and it is able to reach up to 364 giga floating point operations per second (GFLOPS)/W efficiency (32-bit precision) on typical convolutional neural network operators.

97 MATHEMATICS AND COMPUTING↗

Mahanaxar: quality of service guarantees in high-bandwidth, real-time streaming data storage

Large radio telescopes, cyber-security systems monitoring real-time network traffic, and others have specialized data storage needs: guaranteed capture of an ultra-high-bandwidth data stream, retention of the data long enough to determine what is 'interesting,' retention of interesting data indefinitely, and concurrent read/write access to determine what data is interesting, without interrupting the ongoing capture of incoming data. Mahanaxar addresses this problem. Mahanaxar guarantees streaming real-time data capture at (nearly) the full rate of the raw device, allows concurrent read and write access to the device on a best-effort basis without interrupting the data capture, and retains data as long as possible given the available storage. It has built in mechanisms for reliability and indexing, can scale to meet arbitrary bandwidth requirements, and handles both small and large data elements equally well. Results from our prototype implementation shows that Mahanaxar provides both better guarantees and better performance than traditional file systems.

97 MATHEMATICS AND COMPUTING↗

Pianola: A script-based I/O benchmark

Script-based I/O benchmarks record the I/O behavior of applications by using an instrumentation library to trace I/O events and their timing. A replay engine can then reproduce these events from the script in the absence of the original application. This type of benchmark reproduces real-world I/O workloads without the need to distribute, build, or run complex applications. However, faithfully recreating the I/O behavior of the original application requires careful design in both the instrumentation library and the replay engine. This paper presents the Pianola script-based benchmarking system, which includes an accurate and unobtrusive instrumentation system and a simple-to-use replay engine, along with some additional utility programs to manage the creation and replay of scripts. We show that for some sample applications, Pianola reproduces the qualitative features of the I/O behavior. Moreover, the overall replay time and the cumulative read and write times are usually within 10% of the values measured for the original applications.

99 GENERAL AND MISCELLANEOUS↗

Accelerating Flash-X Simulations with Asynchronous I/O

Most high-fidelity physics simulation codes, such as Flash-X, need to save intermediate results (checkpoint files) to restart or gain insights into the evolution of the simulation. These simulation codes save such intermediate files synchronously, where computation is stalled while the data is written to storage. Depending on the problem size and computational requirements, this file write time can be a substantial portion of the total simulation time. In order to hide the I/O latency of checkpointing, asynchronous I/O methods have been introduced. These methods use background threads for performing I/O while the main threads continue with the simulation. The usage of background threads can compete for resources on the node as well as with communication. In this paper, we evaluate the overheads and the overall benefit of asynchronous I/O in HDF5 to simulations. Results from real-world high-fidelity simulations on the Summit supercomputer show that I/O operation is overlapped with application communication or computation or both, effectively hiding some or all of the I/O latency. Our evaluation shows that while using asynchronous I/O adds overhead to the application, the I/O time reduction is more significant, resulting in overall up to 1.5X performance speedup.

Jain, Rajeev↗

The Design, Deployment, and Evaluation of the CORAL Pre-Exascale Systems

CORAL, the Collaboration of Oak Ridge, Argonne and Livermore, is fielding two similar IBM systems, Summit and Sierra, with NVIDIA GPUs that will replace the existing Titan and Sequoia systems. Summit and Sierra are currently ranked No. 1 and No. 3, respectively on the Top500 list. We discuss the design and key differences of the systems. Our evaluation of the systems highlights the following. Applications that fit in HBM see the most benefit and may prefer more GPUs; however, for some applications, the CPU-GPU bandwidth is more important than the number of GPUs. The node-local burst buffer scales linearly, and can achieve a 4X improvement over the parallel file system for large jobs; smaller jobs, however, may benefit from writing directly to the PFS. Finally, several CPU, network and memory bound analytics and GPU-bound deep learning codes achieve up to a 11X and 79X speedup/node, respectively over Titan.

Vazhkudai, Sudharshan↗

GPU Lifetimes on Titan Supercomputer: Survival Analysis and Reliability

The Cray XK7 Titan was the top supercomputer system in the world for a long time and remained critically important throughout its nearly seven year life. It was an interesting machine from a reliability viewpoint as most of its power came from 18,688 GPUs whose operation was forced to execute three rework cycles, two on the GPU mechanical assembly and one on the GPU circuitboards. We write about the last rework cycle and a reliability analysis of over 100,000 years of GPU lifetimes during Titan’s 6-year-long productive period. Using time between failures analysis and statistical survival analysis techniques, we find that GPU reliability is dependent on heat dissipation to an extent that strongly correlates with detailed nuances of the cooling architecture and job scheduling. We describe the history, data collection, cleaning, and analysis and give recommendations for future supercomputing systems. We make the data and our analysis codes publicly available.

Ostrouchov, George↗

An Efficient Checkpointing System for Large Machine Learning Model Training

As machine learning models increase in size and complexity rapidly, the cost of checkpointing in ML training became a bottleneck in storage and performance (time). For example, the latest GPT-4 model has massive parameters at the scale of 1.76 trillion. It is highly time and storage consuming to frequently writes the model to checkpoints with more than 1 trillion floating point values to storage. This work aims to understand and attempt to mitigate this problem. First, we characterize the checkpointing interface in a collection of representative large machine learning/language models with respect to storage consumption and performance overhead. Second, we propose the two optimizations: i) A periodic cleaning strategy that periodically cleans up outdated checkpoints to reduce the storage burden; ii) A data staging optimization that coordinates checkpoints between local and shared file systems for performance improvement.

machine learning, artificial intelligence↗

New Total-Ionizing-Dose Resistant Data Storing Technique for NAND Flash Memory

This paper describes a new non-charge-based data storing technique in NAND flash memory called watermark that encodes read-only data in the form of physical properties of flash memory cells. Unlike traditional charge-based data storing method in flash memory, the proposed technique is resistant to total ionizing dose (TID) effects. To evaluate its resistance to irradiation effects, we analyze data stored in several commercial single-level-cell (SLC) flash memory chips from different vendors and technology nodes. These chips are irradiated using a Co-60 gamma-ray source array for up to 100 krad(Si) at Sandia National Laboratories. Experimental evaluation performed on a flash chip from Samsung shows that the intrinsic bit error rate (BER) of watermark increases from 0.8% for TID = 0 krad(Si) to 1% for TID = 100 krad(Si). Conversely, the BER of charge-based data stored on the same chip increases from 0% at TID = 0 krad(Si) to 1.5% at TID = 100 krad(Si). Overall, the results imply that the proposed technique may potentially offer significant improvements in data integrity relative to traditional charge-based data storage for very high radiation (TID > 100 krad(Si)) environments. These gains in data integrity relative to the charge-based data storage are useful in radiation-prone environments, but they come at the cost of increased write times and higher BERs before irradiation.

36 MATERIALS SCIENCE↗

Design, Fabrication, and Characterization of Near-Field Apertures for 1 Tbit/in$^{2}$ Areal Density

Today, conventional magnetic recording schemes are coming to an end because of the superparamagnetic limit. Heat-assisted magnetic recording (HAMR) may ultimately extend data densities beyond 1 TB/in. HAMR systems utilize the phenomenon during which the magnetic properties of the recording media could be locally modified via heating (optionally, by an optical source in the near field) to temperature in the vicinity of the Curie value of the media material. As a result, heat induced by the optical source can temporarily reduce the magnetic coercivity of high anisotropy material to a level attainable by the magnetic writing head, thus making it feasible to record on relatively small ultra-high anisotropy (and thermally stable) grains, consequently enhancing the areal density dramatically. The key challenge is to develop a near-field transducer capable of delivering over 50 nWinto a spot diameter of 30 nm. Traditional fiber schemes are barely capable of 0.1 nW. To resolve the issue, a laser diode could be placed with the emitting edge only a few nanometers away from the recording media. The light can propagate through a nanoaperture on the surface of an aluminum-coated emitting edge. This paper will present an experimental study of recording characteristics of various near-field transducers fabricated via focused ion beam (FIB). To count the number of photons emitted in the near field, a scanning near-field optical microscopy system has been implemented. The experiments indicate that the FIB-fabricated transducers could deliver power of over a few microwatt into a 30-nm spot

77 NANOSCIENCE AND NANOTECHNOLOGY↗