SEARCH · Engineering Papers

Results for “STORAGE”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,117 records · Page 62

All sources NASA NTRS DOE OSTI

DOE OSTI2023

Geological Thermal Energy Storage Using Solar Thermal and Carnot Batteries: Techno-Economic Analysis: Preprint

Energy storage is increasingly necessary as Variable Renewable Energy (VRE) technologies are deployed. Seasonal energy storage can shift energy generation from the summer to the winter, but these technologies must have extremely large energy capacities and low costs. Geological Thermal energy storage (GeoTES) is proposed as a solution for long-term energy storage. Excess thermal energy can be stored in permeable reservoirs such as aquifers and depleted hydrocarbon reservoirs for several months. In this article, we describe a techno-economic model that has been developed to evaluate GeoTES systems. The models are developed by combining the output of specialist models which enables the performance and cost of both the subsurface and surface systems to be captured. Off-design models are developed so that the performance can be evaluated at each hour of the year. GeoTES can be charged with two different energy sources: (1) concentrating solar thermal and (2) renewable electricity using heat pumps (henceforth known as a "Carnot Battery"). The stored thermal energy can be used to generate electricity and - uniquely - also directly produce heat that can be used by industrial processes. Furthermore, Carnot-Battery-GeoTES can also be used to form a cold storage reservoir. Preliminary results that quantify the technical and economic performance of these two GeoTES systems are presented.

carnot battery↗

DOE OSTI2016

Alternative Canister Processing Operations at an Interim Storage Facility - 16617

The US Department of Energy (DOE) is laying the groundwork for implementing interim storage of spent nuclear fuel (SNF) in response to recommendations by the Blue Ribbon Commission on America's Nuclear Future in their report to the Secretary of Energy published January 2012. These efforts include planning for a Pilot Interim Storage Facility (ISF) and a Larger ISF. In March 2014, the DOE employed the CB and I team to evaluate design alternatives for cask handling operations and storing SNF at an ISF. The report, 'Generic Design Alternatives for Dry Storage of Used Nuclear Fuel', (Alternatives Report) evaluates different aspects of an ISF, including a study on alternative approaches for processing SNF canisters such as different cask handling methods and configurations which might be used to transfer canisters from shipping casks into storage locations at an ISF. The purpose of this paper is to report on the development of alternative cask handling operations that have been evaluated for the receipt and storage of the SNF at an initial Pilot ISF and a Larger ISF. This design study is relevant for host communities considering near-term development of a Pilot ISF. DOE-sponsored system analyses indicate that a processing rate of up to 3,000 MTU/yr (approximately 230 to 300 canisters) would be needed for a single Larger ISF (less if more than one ISF were developed) to remove SNF from commercial nuclear plants in a reasonable amount of time. The study evaluated different canister processing operation methodologies in order to reduce the complexity, cost, and personnel exposure associated with processing operations at an ISF. The conclusions from this design study could help to inform the selection of future canister processing operations for an ISF. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2018

Systems and Tools to Facilitate Aging Management Inspections of Used Nuclear Fuel Dry Storage Systems - 18382

The Used Nuclear Fuel (UNF) dry storage systems (DSSs) are currently used at almost all the nuclear plants in the USA for the storage of their UNF as the spent fuel pools are running out of capacity to store any additional discharged fuel. Even though these DSSs are supposed to be at the reactor site for a short amount of time, they are expected to be at the reactor sites for an extended duration due to the unavailability of Interim Storage Facilities (ISF) or a permanent disposal sites like Yucca Mountain. Maintaining the integrity of the DSS canisters and storage overpacks so they can safely protect the fuel and maintain the integrity of UNF is of utmost importance during storage at reactor site ISFSI or at the ISF. Comprehensive and robust Aging Management Program (AMP) is an important aspect for DSSs to satisfy these requirements. In order to address this industry need, TN Americas have developed an enhanced NUHOMS{sup R} DSS licensed by NRC under 10 CFR Part 72 regulations and a comprehensive AMP development which includes prevention, inspection, monitoring and mitigation of any corrosion or degradation of DSS components. For DSC and HSM inspection, two remotely operating delivery systems have been developed using non-destructive examination (NDE) methods including robotic crawler and inspection ring system. This paper describes details of these developments and status of their deployment for implementation at various utility ISFSI sites. TN Americas completed the detailed design of the DSC inspection system and conducted functional tests of the inspection system on a DSC mockup and the system is ready for deployment at any NUHOMS{sup R} ISFSI site. TN Americas is committed to offer the most robust and comprehensive AMP systems, tools and services to the industry to assure the safety of the deployed DSS at the utility sites. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2018

Monitoring and Modeling Caprock Integrity at the In Salah Carbon Dioxide Storage Site, Algeria

The In Salah gas development project, initiated in 2004, involves the long-term storage of waste carbon dioxide associated with natural gas production at several central Algerian fields. Rather then vent the extracted CO 2 into the atmosphere the operators, BP, Sonatrach, and Statoil chose to have the gas compressed, dehydrated, transported, and injected into a deep saline aquifer leg of the Krechba gas reservoir (Matheison et al. 2011). For this effort three horizontal state-of-the-art injection wells were drilled to a depth of 1500-1800m, using geo-steering technology to follow the roughly 20m thick reservoir. To maximize the injectivity, the wells were oriented perpendicular to the direction of maximum stress, the dominant fracture direction. Long-term plans were to store up to 17 million tonnes of carbon dioxide during the life of the program (Matheison et al. 2010). The project at In Salah was a pioneering effort in several respects and involved the investment of over 100 million dollars by the member companies. One feature that distinguishes In Salah from several other industrialscale CO 2 sequestration projects, such as Sleipner (Chadwick et al. 2012), Weyburn (White et al. 2011), Decatur (Kaven et al. 2014), and Aquastore (Worth et al. 2014) is the generally smaller storage volume in terms of overall reservoir thickness and lower average porosities of around 15%. With the exception of Weyburn, these other sites are characterized by generally thick sequences of high porosity sandstones. Therefore, one might expect that the pressure buildup near the wells could be larger at In Salah, with consequences for reservoir deformation and caprock strain. The role of fractures in providing storage and increasing reservoir permeability is one factor that would likely impact the pressure distribution around the well. However, the detailed effect of existing fractures on flow was not established prior to the injection at In Salah. The unique features of the In Salah storage site point to the importance of wellbore, geophysical, and geochemical monitoring to ensure that the integrity of the storage complex is maintained both during and after injection. In 2005 a Joint Industry Project (JIP) was set up, in an effort to use various geophysical, geochemical, and production techniques to monitor the fate of the injected carbon dioxide (Matheison et al. 2011). Here in this paper we describe the efforts made to monitoring the fate of the injected carbon dioxide, the state of the caprock, the estimated outcome of the storage effort, and some of the uncertainties that remain.

58 GEOSCIENCES↗

DOE OSTI2021

Radial storage efficiency for CO 2 injection: Quantifying effectiveness of local flow control methods

Basin-scale geologic carbon sequestration will require hundreds of injection wells, each of which has costs related to property rights and regulatory requirements that correlate with the areal size of the carbon dioxide plume. These surface-footprint-related issues motivate maximizing storage efficiency radially away from each well. Radial storage efficiency (RSE) is defined here as the ratio of the volumetrically weighted carbon dioxide free-phase saturation within a given radius away from the injection well to the available pore space within that same radius. Maximizing RSE effectively minimizes the radial extent of the CO 2 plume. Optimizing RSE around individual wells requires local flow control injection strategies that can increase the uniformity of the filling of pore space around the well over the entire length of the perforated injection interval despite differences in local formation transmissivity. The goal of uniform filling of the storage reservoir starting from the well outward is to maximize carbon dioxide sweep and trapping locally outward from the well in all of the layers of the storage region before buoyancy forces predominate and drive carbon dioxide upward where it will spread laterally under lower-permeability layers. Example simulations of carbon dioxide injection into a layered storage system with and without local flow control are presented to show the advantage of uniformly filling all layers and how RSE can be used to quantify storage efficiency for the two different injection approaches. Published 2021. This article is a U.S. Government work and is in the public domain in the USA.

03 NATURAL GAS↗

DOE OSTI2018

Immobile Pore-Water Storage Enhancement and Retardation of Gas Transport in Fractured Rock

The effect of immobile pore-water on gas transport in fractured rock has implications for numerical modeling of soil vapor extraction, methane leakage, gaseous CO 2 leakage from sequestration operations, radionuclide transport from underground nuclear explosions, and nuclear waste disposal. While the ability for immobile pore-water storage to effect gas transport has been recognized in the past, the details and specific scenarios leading to enhanced, retarded, or unaffected gas transport have not been explored. In this paper, we performed numerical investigations into the enhancement and retardation of gas transport due to immobile pore-water storage in order to identify implications for gas transport applications. To do this, we developed a numerical approach to model gas transport with a single-phase flow solution coupled to the advection–dispersion equation modified to account for immobile pore-water storage. Other than the immobility of pore water, the formulation contains all other physics included in two-phase formulations (advective and diffusive gas transport in fractures and rock matrix and dissolution in immobile pore water). The assumption of immobile pore water is valid here since for many applications involving transport of soluble gases in fractured rock, the rate of aqueous transport is insignificant compared to gas transport. We verify our modeling approach with analytical solutions of: (1) 1D gas diffusion, (2) 1D gas advection, (3) barometric pumping of a fracture, and (4) gas transport with uniform fracture flow and diffusion into the matrix. We account for pore-water storage in our model by implementing a kinetic formulation of gas dissolution wherein the dissolved (aqueous) phase is considered an immobile constituent. Using this formulation, we model the effect of dissolution rate and saturation on the retardation of gas transport during pure diffusion and pure advection. We also demonstrate that although it is commonly believed that pore-water storage will always enhance gas transport in fractures during oscillatory flow (e.g., during reversing pressure gradients such as barometric pumping cycles), our simulations indicate that this may not always be the case. Our numerical investigations indicate that scenarios with lower effective diffusion coefficients ( ⪅ 10 − 5  m 2 / s ) and lower dissolution coefficients (i.e., the dissolution diffusion coefficient) ( ⪅ 10 − 11  m 2 / s ) will result in enhanced gas transport. Other combinations of gaseous diffusion and dissolution coefficients result in delayed gas transport or insignificant effect on gas transport. Based on these results, and given the fact that the free-air diffusion coefficients for many gases are slightly above 10 − 5  m 2 / s (near the boundary of enhanced/non-enhanced gas transport), tortuosity would have to be significant for gas transport enhancement to occur. Similarly, if we consider the free-water diffusion coefficient of gases (around 10 − 9  m 2 / s for many gases of interest) to be a maximum bound for the effective dissolution coefficients (i.e., the rate of diffusion into the water away from the air/water interface will limit the dissolution rate), the effective dissolution coefficient would have to be at least 2 orders of magnitude less for enhanced gas transport to occur. Otherwise, pore-water storage will delay or have negligible effect on the gas transport. The results also indicate that the assumption of instantaneous equilibrium, often invoked in numerical codes for dissolution processes (i.e., dissolution coefficient is effectively infinite), will fail to accurately capture important details of soluble gas transport. Finally, the implication of our results is that gas breakthrough times in fractured rock during oscillatory flow may be highly dependent on the effective gaseous diffusion and dissolution coefficients.

58 GEOSCIENCES↗

DOE OSTI2022

Economic analysis of CCUS: Accelerated development for CO 2 EOR and storage in residual oil zones under the context of 45Q tax credit

Residual oil zones (ROZ) undergoing CO 2 Enhanced Oil Recovery (CO 2 -EOR) may benefit from specific strategies to maximize their value. We evaluated several strategies for producing from a Permian Basin, West Texas, USA field’s ROZ. This ROZ lies below the main pay zone (MPZ) of the field. Such brownfield ROZs occur in the Permian Basin and elsewhere. Since brownfield ROZs are hydraulically connected to the MPZs, development sequences and schemes influence oil production, CO 2 storage, and net present value (NPV). We conducted economic assessments of various CO 2 injection/production schemes in the stacked ROZ-MPZ reservoir based on flow simulations of a high-resolution geocellular model built from wireline logs and core data and calibrated through production history matching. Flow simulations of water alternating gas (WAG) injection, such as switching injection from the MPZ to the ROZ and commingled production, were studied. Simulation results showed that simultaneous CO 2 injection into the MPZ and ROZ lead to the largest oil production and, generally, the largest NPV. If instead, CO 2 was simultaneously injected into the MPZ and ROZ, then into the ROZ alone, this maximized CO 2 storage. CO 2 storage can be used as a tax credit under the Internal Revenue Code, Section 45Q. Storage performance depends on the development approach and WAG ratio. Developing the ROZ increased storage compared to only producing from the MPZ. The WAG ratio to maximize oil production did not always yield the largest NPV. These findings are potentially applied to other Brownfield ROZs, which are common below San Andres reservoirs in the Permian Basin and other basins. ROZ development can increase oilfields’ NPV and carbon storage potential. Our study can serve as an analog for similar reservoirs. Here this work provides valuable insights into the further optimization of brownfield ROZ development and information for operators to plan to develop stacked ROZ-MPZ reservoirs.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2018

The splicing factor transformer2 (tra2) functions in the Drosophila fat body to regulate lipid storage

Highlights: • Transformer2 (tra2) is a splicing protein whose metabolic functions are unknown. • Tra2 functions in the Drosophila fat body to modulate triglyceride levels. • Decreasing tra2 in the fat body promotes starvation resistance. • Tra2 alters CPT1 splicing in the fly fat body. Excess nutrients are stored as triglycerides mainly in the adipose tissue of an animal and these triglycerides are located in structures called lipid droplets. Previous genome-wide RNAi screens in Drosophila cells identified splicing factors as playing a role in lipid droplet formation. Our lab has recently identified the SR protein, 9G8, as an important factor in fat storage as decreasing its levels results in augmented triglyceride storage in the fat body. Previous in vitro studies have implicated 9G8 in the regulation of splicing of the sex determination gene doublesex (dsx) by binding to transformer (tra) and transformer2 (tra2); however, any function of these sex determination proteins in regulating metabolism is unknown. In this study, we have uncovered a role of tra2 to regulate fat storage in vivo. Inducing tra2{sup dsRNA} in the adult fat body resulted in an increase in triglyceride levels but had no effect on glycogen storage. Consistent with the triglyceride phenotype, tra2 knockdown flies lived longer under starvation conditions. In addition, this increase in triglycerides is due to more fat storage per cell and not an increase in the number of fat cells. Interestingly, the splicing of CPT1, an enzyme involved in the breakdown of lipids, was altered in flies with decreased tra2. The less-catalytically active isoform of CPT1 accumulated in tra2{sup dsRNA} flies suggesting a decrease in lipid breakdown, which is consistent with the increased triglyceride levels observed in these flies. Together, these results suggest a link between mRNA splicing, sex determination and lipid metabolism and may provide insight into the mechanisms underlying tissue-specific splicing and nutrient storage.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2017

Integrated CO 2 Storage and Brine Extraction

Carbon dioxide (CO 2 ) capture, utilization, and storage (CCUS) can reduce CO 2 emissions from fossil fuel power plants by injecting CO 2 into deep saline aquifers for storage. CCUS typically increases reservoir pressure which increases costs, because less CO 2 can be injected, and risks such as induced seismicity. Extracting brine with enhanced water recovery (EWR) from the CO 2 storage reservoir can manage and reduce pressure in the formation, decrease the risks linked to reservoir overpressure (e.g., induced seismicity), increase CO 2 storage capacity, and enable CO 2 plume management. We modeled scenarios of CO 2 injection with EWR into the Rock Springs Uplift (RSU) formation in southwest Wyoming. The Finite Element Heat and Mass Transfer Code (FEHM) was used to model CO 2 injection with brine extraction and the corresponding increase in pressure within the RSU. We analyzed the model for pressure management, CO 2 storage, CO 2 saturation, and brine extraction due to the quantity and location of brine extraction wells. The model limited CO 2 injection to a constant pressure increase of two MPa at the injection well with and without extracting brine at hydrostatic pressure. Finally, we found that brine extraction can be used as a technical and cost-effective pressure management strategy to limit reservoir pressure buildup and increase CO 2 storage associated with a single injection well.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2019

Characterization of CO 2 storage and enhanced oil recovery in residual oil zones

Residual oil zones (ROZs) are reservoirs in which oil is swept over geologic time period and exists at residual saturation. The oil in such reservoirs cannot be commercially exploited using conventional oil recovery methods as the oil exists at residual oil saturation. Instead, enhanced oil recovery methods such as CO 2 injection are required. Recently, ROZs have been increasingly studied as potential CO 2 storage targets. In spite of increased interest in ROZs, there are significant gaps in the knowledge of parameters and processes that impact CO 2 storage and oil recovery. In this work, we identify key geologic and operational characteristics that affect CO 2 storage capacity and oil recovery potential by performing Monte Carlo simulations and sensitivity analysis. In addition to CO 2 storage capacity, we also characterize the long-term CO 2 fate in ROZs. The distinction of CO 2 storage in ROZs from conventional oil reservoirs and saline aquifers are also characterized. Furthermore, predictive models based on machine learning techniques are developed to estimate CO 2 storage and oil production potentials for ROZs. The applicability of the predictive models is demonstrated for five ROZs in the Permian Basin.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2026

Modeling the Potential Impact of Storage on the US Power Sector in a Multisector Dynamic Context

The electric power sector is expected to grow in size, importance, and complexity around the world as economies expand and electric supply technologies and demand patterns evolve in significant ways. At the same time the electric power sector may see substantial increases in VRE generating technologies which can add variability and uncertainty to the diurnal and seasonal profile of electric supply. With these forces in play, the emergence of modular, flexible electricity storage technologies may have profound impacts on the operation of electric power systems. Here we reduce a storage modeling gap in MSD models by incorporating grid-based electricity storage into the electric sector dynamics of the Global Change Analysis Model-USA (GCAM-USA) with improved power sector representation. We find a potentially significant role for storage technologies in the future of the U.S. power system, with storage capacity ranging from 7.7 to 14.7 GW in 2050 and 13.9 to 31.6 GW in 2100 across several techno-economic scenarios. We also find that storage can help to smooth variability in residual load arising from evolving electricity demands and the introduction of high shares of VRE to the electric grid. This reduces reliance on high-cost peaking generators, improves system-wide capacity factors, and limits curtailment of VRE.

Patel, Pralit L.↗

DOE OSTI2023

Numerical evaluation of photosensitive tracers as a strategy for separating surface and subsurface transient storage in streams

For this work, we numerically evaluated photosensitive tracers as a potential strategy for separating the effects of surface and hyporheic storage zones (SSZs and HSZs, respectively) on stream corridor transport. Correctly separating HSZ and SSZ effects is critical to estimating the hydro-biogeochemical function of a stream because HSZs and SSZs expose solutes to significantly different biogeochemical conditions, like sunlight exposure, microbial processes, and oxygen availability. Our numerical experiments used a multiscale river-corridor transport model implemented in the ATS code, which accommodates multiple storage zones with distinct travel time distributions and biogeochemical reactions. For parameter inferences, we used Bayesian inverse modeling. We found that breakthrough curves for photo-decaying tracers from day and night injection can delineate surface and hyporheic transient storage contributions, but only when interpreted jointly through a two-storage zone model. Numerical experiments that used only daytime injection or interpreted breakthrough curves with a single storage zone model yielded good fit to breakthrough curves, but parameter estimates were biased and controlling processes misattributed, examples of good model fits for the wrong reasons. Using those biased parameter estimates in reactive transport simulations resulted in significantly different projections of denitrification, which underscores the potential for stream function to be mischaracterized if tracer tests are interpreted through an inappropriately simplified model for transient storage. More generally, this study highlights the role of modeling in evaluating the experimental design and identifying the potential of system mischaracterization and its implications.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2014

Spatial and temporal modeling of sub- and supercritical thermal energy storage

This paper describes a thermodynamic model that simulates the discharge cycle of a single-tank thermal energy storage (TES) system that can operate from the two-phase (liquid-vapor) to supercritical regimes for storage fluid temperatures typical of concentrating solar power plants. State-of-the-art TES design utilizes a two-tank system with molten nitrate salts; one major problem is the high capital cost of the salts (International Renewable Energy Agency, 2012). The alternate approach explored here opens up the use of low-cost fluids by considering operation at higher pressures associated with the two-phase and supercritical regimes. The main challenge to such a system is its high pressures and temperatures which necessitate a relatively high-cost containment vessel that represents a large fraction of the system capital cost. To mitigate this cost, the proposed design utilizes a single-tank TES system, effectively halving the required wall material. A single-tank approach also significantly reduces the complexity of the system in comparison to the two-tank systems, which require expensive pumps and external heat exchangers. A thermodynamic model is used to evaluate system performance; in particular it predicts the volume of tank wall material needed to encapsulate the storage fluid. The transient temperature of the tank is observed to remain hottest at the storage tank exit, which is beneficial to system operation. It is also shown that there is an optimum storage fluid loading that generates a given turbine energy output while minimizing the required tank wall material. Overall, this study explores opportunities to further improve current solar thermal technologies. The proposed single-tank system shows promise for decreasing the cost of thermal energy storage. (C) 2014 Elsevier Ltd. All rights reserved.

Tse, LA↗

DOE OSTI2025

Hydrogen Storage with Aluminum Formate, ALF: Experimental, Computational, and Technoeconomic Studies

Long-duration storage of hydrogen is necessary for coupling renewable H2 with stationary fuel cell power applications. In this presentation, I will discuss how aluminum formate, Al(HCOO)3 (ALF), which adopts an ReO3-type structure, is shown to have remarkable H2 storage performance at non-cryogenic (> 120 K) temperatures and low pressures. The most promising performance of ALF is found between 120 K and 160 K and at 10 bar to 20 bar. The talk will cover and illustrate the H2 adsorption performance of ALF over the 77 K to 296 K temperature range using gas isotherms, in situ neutron powder diffraction, and DFT calculations, as well as technoeconomic analysis (TEA), illustrating ALF’s competitive performance for long-duration storage versus compressed hydrogen and leading metal–organic frameworks. In the TEA, it is shown that ALF’s storage capacity, when combined with a temperature/pressure swing process, has advantages versus compressed H2 at a fraction of the pressure (15 bar versus 350 bar). Given ALF’s performance in the 10 bar to 20 bar regime under moderate cooling, it is particularly promising for use in safe storage systems serving fuel cells, and is currently the only MOF that works in this moderate temperature range/ low pressure regime to be cost competitive with compressed H2 gas for large scale H2 storage.[1]

Chemistry↗

DOE OSTI2015

Mean PB To Failure - Initial results from a long-term study of disk storage patterns at the RACF

The RACF (RHIC-ATLAS Computing Facility) has operated a large, multi-purpose dedicated computing facility since the mid-l 990's, serving a worldwide, geographically diverse scientific community that is a major contributor to various HEPN projects. A central component of the RACF is the Linux-based worker node cluster that is used for both computing and data storage purposes. It currently has nearly 50,000 computing cores and over 23 PB of storage capacity distributed over 12,000+ (non-SSD) disk drives. The majority of the 12,000+ disk drives provide a cost-effective solution for dCache/XRootD-managed storage, and a key concern is the reliability of this solution over the lifetime of the hardware, particularly as the number of disk drives and the storage capacity of individual drives grow. We report initial results of a long-term study to measure lifetime PB read/written to disk drives in the worker node cluster. We discuss the historical disk drive mortality rate, disk drive manufacturers' published MPTF (Mean PB to Failure) data and how they are correlated to our results. The results help the RACF understand the productivity and reliability of its storage solutions and have implications for other highly-available storage systems (NFS, GPFS, CVMFS, etc) with large I/0 requirements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2018

ARCHIE: Data Analysis Acceleration with Array Caching in Hierarchical Storage

© 2018 IEEE. Scientific data analysis typically involves reading massive amounts of data that was generated by simulations, experiments, and observations. Performance of reading such large volumes of data from disk-based file systems is often poor because of the slow and mechanical components in the disks. Recent supercomputing systems are adding non-volatile storage layers in a hierarchy to handle the performance gap between fast main memory and slow disk-based storage. Software libraries for managing this hierarchy not only need efficient reading of data but also reduce user-involvement for cross-layer data movement. Furthermore, these libraries need to support array data access patterns into hierarchical storage management as scientific data is often organized in array-based data structures. Existing software typically manage individual storage layers requiring significant manual process in moving data among them. In this paper, we introduce a new array caching in hierarchical storage (ARCHIE) to accelerate array data analysis in a seamless fashion. ARCHIE evaluates array access patterns and prefetches data with array semantics between storage layers. Our evaluation shows that ARCHIE outperforms state-of-the-art file systems, i.e., Lustre and DataWarp, on a production supercomputing system by up to 5.8× in accessing data by scientific analysis applications.

Dong, Bin↗

DOE OSTI2018

Toward Transparent Data Management in Multi-layer Storage Hierarchy for HPC Systems

Upcoming exascale high performance computing (HPC) systems are expected to comprise multi-tier storage hierarchy, and thus will necessitate innovative storage and I/O mechanisms. Traditional disk and block-based interfaces and file systems face severe challenges in utilizing capabilities of storage hierarchies due to the lack of hierarchy support and semantic interfaces. Object-based and semantically-rich data abstractions for scientific data management on large scale systems offer a sustainable solution to these challenges. Such data abstractions can also simplify users involvement in data movement. Here, we take the first steps of realizing such an object abstraction and explore storage mechanisms for these objects to enhance I/O performance, especially for scientific applications. We explore how an object-based interface can facilitate next generation scalable computing systems by presenting the mapping of data I/O from two real world HPC scientific use cases: a plasma physics simulation code (VPIC) and a cosmology simulation code (HACC). Our storage model stores data objects in different physical organizations to support data movement across layers of memory/storage hierarchy. Our implementation sclaes well to 16K parallel processes, and compared to the state of the art, such as MPI-IO and HDF5, our object-based data abstractions and data placement strategy in multi-level storage hierarchy achieves up to 7 X I/O performance improvement for scientific data.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

Data Jockey: Automatic Data Management for HPC Multi-Tiered Storage Systems

We present the design and implementation of Data Jockey, a data management system for HPC multi-tiered storage systems. As a centralized data management control plane, Data Jockey automates bulk data movement and placement for scientific workflows and integrates into existing HPC storage infrastructures. Data Jockey simplifies data management by eliminating human effort in programming complex data movements, laying datasets across multiple storage tiers when supporting complex workflows, which in turn increases the usability of multitiered storage systems emerging in modern HPC data centers.Specifically, Data Jockey presents a new data management scheme called “goal driven data management” that can automatically infer low-level bulk data movement plans from declarative high-level goal statements that come from the lifetime of iterative runs of scientific workflows. While doing so, Data Jockey aims to minimize data wait times by taking responsibility for datasets that are unused or to be used, and aggressively utilizing the capacity of the upper, higher performant storage tiers.We evaluated a prototype implementation of Data Jockey under a synthetic workload based on a year’s worth of Oak Ridge Leadership Computing Facility’s (OLCF) operational logs. Our evaluations suggest that Data Jockey leads to higher utilization of the upper storage tiers while minimizing the programming effort of data movement compared to human involved, per-domain adhoc data management scripts.

Shin, Woong↗