Search NASA⌕ Search

SEARCH · Search NASA

Results for “storage 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 343 records · Page 19

Reliability Lessons Learned From GPU Experience With The Titan Supercomputer at Oak Ridge Leadership Computing Facility

The high computational capability of graphics processing units (GPUs) is enabling and driving the scientific discovery process at large-scale. The world’s second fastest supercomputer for open science, Titan, has more than 18,000 GPUs that computational scientists use to perform scientific simu- lations and data analysis. Understanding of GPU reliability characteristics, however, is still in its nascent stage since GPUs have only recently been deployed at large-scale. This paper presents a detailed study of GPU errors and their impact on system operations and applications, describing experiences with the 18,688 GPUs on the Titan supercom- puter as well as lessons learned in the process of efficient operation of GPUs at scale. These experiences are helpful to HPC sites which already have large-scale GPU clusters or plan to deploy GPUs in the future.

Tiwari, Devesh↗

Large-scale compute-intensive analysis via a combined in-situ and co-scheduling workflow approach, In: SC '15 Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

Large-scale simulations can produce hundreds of terabytes to petabytes of data, complicating and limiting the efficiency of workflows. Traditionally, outputs are stored on the file system and analyzed in post-processing. With the rapidly increasing size and complexity of simulations, this approach faces an uncertain future. Trending techniques consist of performing the analysis in-situ, utilizing the same resources as the simulation, and/or off-loading subsets of the data to a compute-intensive analysis system. We introduce an analysis framework developed for HACC, a cosmological N-body code, that uses both in-situ and co-scheduling approaches for handling petabyte-scale outputs. We compare different analysis set-ups ranging from purely off-line, to purely in-situ to in-situ/co-scheduling. The analysis routines are implemented using the PISTON/VTK-m framework, allowing a single implementation of an algorithm that simultaneously targets a variety of GPU, multi-core, and many-core architectures.

Computer Science↗

Performance of a distributed superscalar storage server

The RS/6000 performed well in our test environment. The potential exists for the RS/6000 to act as a departmental server for a small number of users, rather than as a high speed archival server. Multiple UniTree Disk Server's utilizing one UniTree Disk Server's utilizing one UniTree Name Server could be developed that would allow for a cost effective archival system. Our performance tests were clearly limited by the network bandwidth. The performance gathered by the LibUnix testing shows that UniTree is capable of exceeding ethernet speeds on an RS/6000 Model 550. The performance of FTP might be significantly faster if asked to perform across a higher bandwidth network. The UniTree Name Server also showed signs of being a potential bottleneck. UniTree sites that would require a high ratio of file creations and deletions to reads and writes would run into this bottleneck. It is possible to improve the UniTree Name Server performance by bypassing the UniTree LibUnix Library altogether and communicating directly with the UniTree Name Server and optimizing creations. Although testing was performed in a less than ideal environment, hopefully the performance statistics stated in this paper will give end-users a realistic idea as to what performance they can expect in this type of setup.

Finestead, Arlan↗

Physics-based seismic hazard analysis on petascale heterogeneous supercomputers, In: SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

We have developed a highly scalable and efficient GPU-based finite-difference code (AWP) for earthquake simulation that implements high throughput, memory locality, communication reduction and communication/computation overlap and achieves linear scalability on Cray XK7 Titan at ORNL and NCSA's Blue Waters system. We simulate realistic 0-10 Hz earthquake ground motions relevant to building engineering design using high-performance AWP. Moreover, we show that AWP provides a speedup by a factor of 110 in key strain tensor calculations critical to probabilistic seismic hazard analysis (PSHA). These performance improvements to critical scientific application software, coupled with improved co-scheduling capabilities of our workflow-managed systems, make a statewide hazard model a goal reachable with existing supercomputers. The performance improvements of GPU-based AWP are expected to save millions of core-hours over the next few years as physics-based seismic hazard analysis is developed using heterogeneous petascale supercomputers.

Computer Science↗

Net-structured Co{sub 3}O{sub 4}/C nanosheet array with enhanced electrochemical performance toward lithium storage

Graphical abstract: - Highlights: • Co{sub 3}O{sub 4}/C nanosheet array on nickel foam was synthesized by hydrothermal method, followed by magnetron sputtering of a carbon layer. • The Co{sub 3}O{sub 4}/C nanosheet array electrode delivers an initial discharge capacity of 1340.4 mAh g{sup −1} at 0.1 C. • After 100 cycles, the capacity retention of the Co{sub 3}O{sub 4}/C is 88.8% of its initial discharge capacity. • The enhanced electrochemical performances are attributed to the porous nanosheet array and the thin carbon layer coated on Co{sub 3}O{sub 4} nanosheets. - Abstract: Freestanding Co{sub 3}O{sub 4}/C nanosheet array growing directly on nickel foam substrate was synthesized using a hydrothermal approach, followed by a direct current magnetron sputtering of a thin carbon layer. The Co{sub 3}O{sub 4} nanosheets with thicknesses of 10–20 nm are interconnected with each other, forming a highly open net-structure. The Co{sub 3}O{sub 4}/C nanosheet array electrode delivers an initial discharge capacity of 1340.4 mAh g{sup −1} at 0.1 C, and maintains a high specific capacity of 912.1 mAh g{sup −1} at room temperature and can still remain 88.8% of its initial discharge capacity at 55 °C at 1 C after 100 cycles. The enhanced electrochemical performances are attributed to the porous nanosheet array and the thin carbon layer coated on Co{sub 3}O{sub 4} nanosheets, which cannot only facilitate Li{sup +} and electron transportation in the electrode, but also improve its structure stability during cycling.

36 MATERIALS SCIENCE↗

Effect of Thermal Storage on the Performance of a Wood Pellet-fired Residential Boiler

Interest in the direct use of biomass for thermal applications as a renewable technology is increasing as is also focus on air pollutant emissions from these sources and methods to minimize the impact. This work has focused on wood pellet-fired residential boilers, which are the cleanest fuel in this category. In the residential application the load varies strongly over the course of a year and a high fraction of the load is typically under 15% of the maximum boiler capacity. Thermal storage can be used even with boilers which have modulation capacity typically to 30% of the boiler maximum. One common pellet boiler was tested at full load and also at the minimum load used in the U.S. certification testing (15%). In these tests the load was steady over the test period. Testing was also done with an emulated load profile for a home in Albany, N.Y. on a typical January, March, and April day. In this case the load imposed on the boiler varied hourly under computer control, based on the modeled load for the example case used. The boiler used has a nominal output of 25 kW and a common mixed hardwood/softwood commercial pellet was used. Moisture content was 3.77%. A dilution tunnel approach was used for the measurement of particulate emissions, in accordance with U.S. certification testing requirements. The test results showed that the use of storage strongly reduces cycling rates under part load conditions. The transients which occur as these boilers cycle contribute to increased particulate emissions and reduced efficiency. The time period of a full cycle at a given load condition can be increased by increasing the storage tank volume and/or increasing the control differential range. It was shown that increasing the period strongly increased the measured efficiency and reduced the particulate emission (relative to the no storage case). The impact was most significant at the low load levels. Storage tank heat loss is shown to be a significant factor in thermal efficiency, particularly at low load. Different methods to measure this heat loss were explored. For one of the tanks evaluated the efficiency loss at the 15% load point was found to be as high as 7.9%. Where storage is used good insulation on the tank, insulation on the piping, and attention to fittings are recommended.

09 BIOMASS FUELS↗

24.77 Pflops on a Gravitational Tree-Code to Simulate the Milky Way Galaxy with 18600 GPUs, In: SC '14: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

We have simulated, for the first time, the long term evolution of the Milky Way Galaxy using 51 billion particles on the Swiss Piz Daint supercomputer with our N-body gravitational tree-code Bonsai. Herein, we describe the scientific motivation and numerical algorithms. The Milky Way model was simulated for 6 billion years, during which the bar structure and spiral arms were fully formed. This improves upon previous simulations by using 1000 times more particles, and provides a wealth of new data that can be directly compared with observations. We also report the scalability on both the Swiss Piz Daint and the US ORNL Titan. On Piz Daint the parallel efficiency of Bonsai was above 95%. The highest performance was achieved with a 242 billion particle Milky Way model using 18600 GPUs on Titan, thereby reaching a sustained GPU and application performance of 33.49 Pflops and 24.77 Pflops respectively.

Computer Science↗

Interface-engineered (AlMnCoNiZn) 3 O 4 @PPy nanocomposites for superior lithium storage: Mechanism and performance

The practical application of graphite anodes in lithium-ion batteries (LIBs) is constrained by low specific capacity (372 mAh g -1 ) and sluggish kinetics. Here, to address these limitations, our present study focuses on high-entropy oxides (HEOs), which offer high theoretical capacity and structural stability. We synthesized spinel-structured (AlMnCoNiZn) 3 O 4 nanoparticles via a solution combustion method and fabricated an (AlMnCoNiZn) 3 O 4 @polypyrrole (PPy) nanocomposite through in-situ polymerization. Our electrochemical tests demonstrate that the PPy modification significantly enhances performance. While the pristine (AlMnCoNiZn) 3 O 4 delivered 445 mAh g −1 after 100 cycles at 100 mA g −1 and 350 mAh g −1 after 1000 cycles at 1000 mA g −1 , the (AlMnCoNiZn) 3 O 4 @PPy composite achieved 695 mAh g −1 after 100 cycles and maintained 675 mAh g −1 after 1000 cycles. Furthermore, the composite improved rate capacity at 1000 mA g −1 from 211 mAh g −1 to 403 mAh g −1 . This work highlights how conductive and flexible polymer modifications can dramatically improve the electrochemical properties of HEOs. The developed (AlMnCoNiZn) 3 O 4 @PPy composite provides a promising direction for designing advanced anodes to meet next-generation energy storage demands.

Anodes material↗

Using cross-layer adaptations for dynamic data management in large scale coupled scientific workflows, In: SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

As system scales and application complexity grow, managing and processing simulation data has become a significant challenge. While recent approaches based on data staging and in-situ/in-transit data processing are promising, dynamic data volumes and distributions, such as those occurring in AMR-based simulations, make the efficient use of these techniques challenging. In this paper we propose cross-layer adaptations that address these challenges and respond at runtime to dynamic data management requirements. Specifically we explore (1) adaptations of the spatial resolution at which the data is processed, (2) dynamic placement and scheduling of data processing kernels, and (3) dynamic allocation of in-transit resources. We also exploit coordinated approaches that dynamically combine these adaptations at the different layers. We evaluate the performance of our adaptive cross-layer management approach on the Intrepid IBM-BlueGene/P and Titan Cray-XK7 systems using Chombo-based AMR applications, and demonstrate its effectiveness in improving overall time-to-solution and increasing resource efficiency.

Computer Science↗

Tehachapi Wind Energy Storage Project - Technology Performance Report #3

The TSP is located at SCE’s Monolith Substation in Tehachapi, California. The 8 MW, 4 hours (32 MWh) BESS is housed in a 6,300 square foot facility and 2 x 4 MW/4.5 MVA smart inverters are on a concrete pad adjacent to the BESS facility. The project will evaluate the capabilities of the BESS to improve grid performance and assist in the integration of large-scale intermittent generation, e.g., wind. Project performance was measured by 13 specific operational uses: providing voltage support and grid stabilization, decreasing transmission losses, diminishing congestion, increasing system reliability, deferring transmission investment, optimizing renewable-related transmission, providing system capacity and resources adequacy, integrating renewable energy (smoothing), shifting wind generation output, frequency regulation, spin/non-spin replacement reserves, ramp management, and energy price arbitrage. Most of the operations either shift other generation resources to meet peak load and other electricity system needs with stored electricity, or resolve grid stability and capacity concerns that result from the interconnection of intermittent generation. SCE also demonstrated the ability of lithium ion battery storage to provide nearly instantaneous maximum capacity for supply-side ramp rate control to minimize the need for fossil fuel-powered back-up generation. The project began in October, 2010 and will continue through December, 2016.

25 ENERGY STORAGE↗

Pushing back the limit of ab-initio quantum transport simulations on hybrid supercomputers, In: SC '15 Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

The capabilities of CP2K, a density-functional theory package and OMEN, a nano-device simulator, are combined to study transport phenomena from first-principles in unprecedentedly large nanostructures. Based on the Hamiltonian and overlap matrices generated by CP2K for a given system, OMEN solves the Schrödinger equation with open boundary conditions (OBCs) for all possible electron momenta and energies. To accelerate this core operation a robust algorithm called SplitSolve has been developed. It allows to simultaneously treat the OBCs on CPUs and the Schrödinger equation on GPUs, taking advantage of hybrid nodes. Our key achievements on the Cray-XK7 Titan are (i) a reduction in time-to-solution by more than one order of magnitude as compared to standard methods, enabling the simulation of structures with more than 50000 atoms, (ii) a parallel efficiency of 97% when scaling from 756 up to 18564 nodes, and (iii) a sustained performance of 15 DP-PFlop/s.

Computer Science↗

In-Situ Feature Extraction of Large Scale Combustion Simulations Using Segmented Merge Trees, In: SC '14: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

The ever increasing amount of data generated by scientific simulations coupled with system I/O constraints are fueling a need for in-situ analysis techniques. Of particular interest are approaches that produce reduced data representations while maintaining the ability to redefine, extract, and study features in a post-process to obtain scientific insights. This paper presents two variants of in-situ feature extraction techniques using segmented merge trees, which encode a wide range of threshold based features. The first approach is a fast, low communication cost technique that generates an exact solution but has limited scalability. The second is a scalable, local approximation that nevertheless is guaranteed to correctly extract all features up to a predefined size. We demonstrate both variants using some of the largest combustion simulations available on leadership class supercomputers. Our approach allows state-of-the-art, feature-based analysis to be performed in-situ at significantly higher frequency than currently possible and with negligible impact on the overall simulation runtime.

Computer Science↗

Tuning inner-layer oxygen functional groups of reduced graphene oxide by potentiostatic oxidation for high performance electrochemical energy storage devices

The electrochemical lithiation/delithiation of oxygen-containing functional groups (OCFGs) of nanocarbon materials, particularly graphene, have attracted intensive interest in recent years. Here, we propose a controllable potentiostatic oxidation approach to tune the OCFGs of as-prepared reduced graphene oxide (rGO) in a carbonate-based electrolyte to improve the specific capacity and rate capability. Furthermore, by X-Ray absorption spectroscopy in total fluorescence yield mode and X-Ray diffraction, we confirm that potentiostatic oxidations generate new OCFGs in the inner-layer of rGO. The content of OCFGs increases as oxidation potential being elevated. Such increasing of OCFGs in quantity significantly enhances the capacity. For instance, the specific capacity of 170.4 mAh g -1 for pristine rGO electrode is increased to 290.5 mAh g -1 after the oxidation at 5.0 V. Finally, we demonstrate that oxidations at moderate potentials can reduce the electrochemical and ohmic polarizations of rGO electrodes without deteriorating diffusion dynamic, thereby improving rate capability. After the optimal oxidation at 4.7 V, rGO electrode exhibits an excellent rate capability, delivering 58.4 mAh g -1 at 20 A g -1 .

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

FRAPCON analysis of cladding performance during dry storage operations

There is an increasing need in the U.S. and around the world to move used nuclear fuel from wet storage in fuel pools to dry storage in casks stored at independent spent fuel storage installations (ISFSI) or interim storage sites. The NRC limits cladding temperature to 400°C while maintaining cladding hoop stress below 90 MPa in an effort to avoid radial hydride reorientation. An analysis was conducted with FRAPCON-4.0 on three modern fuel designs with three representative used nuclear fuel storage temperature profiles that peaked at 400 °C. Results were representative of the majority of U.S. LWR fuel. They conservatively showed that hoop stress remains below 90 MPa at the licensing temperature limit. Results also show that the limiting case for hoop stress may not be at the highest rod internal pressure in all cases but will be related to the axial temperature and oxidation profiles of the rods at the end of life and in storage.

Dry Storage↗

Best Practices and Lessons Learned from Deploying and Operating Large-Scale Data-Centric Parallel File Systems, In: SC '14: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

The Oak Ridge Leadership Computing Facility (OLCF) has deployed multiple large-scale parallel file systems (PFS) to support its operations. During this process, OLCF acquired significant expertise in large-scale storage system design, file system software development, technology evaluation, benchmarking, procurement, deployment, and operational practices. Based on the lessons learned from each new PFS deployment, OLCF improved its operating procedures, and strategies. This paper provides an account of our experience and lessons learned in acquiring, deploying, and operating large-scale parallel file systems. We believe that these lessons will be useful to the wider HPC community.

Computer Science↗