Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “RACKS”

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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
NASA NTRS2009

T-Slide Linear Actuators

T-slide linear actuators use gear bearing differential epicyclical transmissions (GBDETs) to directly drive a linear rack, which, in turn, performs the actuation. Conventional systems use a rotary power source in conjunction with a nut and screw to provide linear motion. Non-back-drive properties of GBDETs make the new actuator more direct and simpler. Versions of this approach will serve as a long-stroke, ultra-precision, position actuator for NASA science instruments, and as a rugged, linear actuator for NASA deployment duties. The T slide can operate effectively in the presence of side forces and torques. Versions of the actuator can perform ultra-precision positioning. A basic T-slide actuator is a long-stroke, rack-and-pinion linear actuator that, typically, consists of a T-slide, several idlers, a transmission to drive the slide (powered by an electric motor) and a housing that holds the entire assembly. The actuator is driven by gear action on its top surface, and is guided and constrained by gear-bearing idlers on its other two parallel surfaces. The geometry, implemented with gear-bearing technology, is particularly effective. An electronic motor operating through a GBDET can directly drive the T slide against large loads, as a rack and pinion linear actuator, with no break and no danger of back driving. The actuator drives the slide into position and stops. The slide holes position with power off and no brake, regardless of load. With the T slide configuration, this GBDET has an entire T-gear surface on which to operate. The GB idlers coupling the other two T slide parallel surfaces to their housing counterpart surfaces provide constraints in five degrees-of-freedom and rolling friction in the direction of actuation. Multiple GB idlers provide roller bearing strength sufficient to support efficient, rolling friction movement, even in the presence of large, resisting forces. T-slide actuators can be controlled using the combination of an off-the-shelf, electric servomotor, a motor angle resolution sensor (typically an encoder or resolver), and microprocessor-based intelligent software. In applications requiring precision positioning, it may be necessary to add strain gauges to the T-slide housing. Existing sensory- interactive motion control art will work for T slides. For open-loop positioning, a stepping motor emulation technique can be used.

Vranish, John↗
NASA NTRS2000

Deformation and Flexibility Equations for Idealized ARIS Umbilicals, Under Planar End-Loading Conditions

The International Space Station (ISS) relies on the Active Rack Isolation System (ARIS) as the central component of an integrated, station-wide strategy to isolate microgravity space-science experiments. ARIS uses electromechanical actuators to isolate an International Standard Payload Rack (ISPR) from disturbances due to the motion of the ISS. Disturbances to microgravity experiments on ARIS-isolated racks are primarily transmitted via the ARIS power and vacuum umbilicals. Recent experimental tests indicate that these umbilicals resonate at frequencies outside the ARIS controller's bandwidth. at levels of potential concern for certain microgravity experiments. Reduction in the umbilical resonant frequencies could help to address this issue. This paper develops equations for the in-plane deflections and flexibilities of an idealized umbilical (thin, flexible, cantilever beam) under end-point, in-plane loading (inclined-force and moment). The effect of gravity is neglected due to the on:orbit application. The analysis assumes an initially straight. cantilevered umbilical with uniform cross-section. which undergoes large deflections with no plastic deformation, such that the umbilical terminus remains in a single quadrant and the umbilical slope changes monotonically. The analysis is applicable to the ARIS power and vacuum umbilicals. under the indicated assumptions.

Hampton, R. David↗
NASA NTRS2010

Overview of the Development of the Temporary Sleep Station Hygiene Liner Aboard the International Space Station

Since the beginning of manned operations aboard the International Space Station (ISS), the crew had performed hygiene activities within the aisle way (the habitable volume, not including the sleep areas) of the ISS. The Crew used wet towels, re-hydrated body soap, and "no-rinse" shampoo to cleanse themselves amongst the stowage and systems hardware, referred to as "racks", even without a designated area to dry the wet items. Performing hygiene in this manner became an accepted method; no isolated location was available to the Crew. After several years of hygiene operations, some of the fabric-covered racks began to grow biological material (generically described as mold) and soon became a Crew health concern. Hygiene has one of the strongest impacts on Crew morale, and mandating changes to the Crew routine would have been met with strong resistance. The answer to the conundrum was to develop a liner to be placed within the Temporary Sleep Station (TeSS), one of the Crew s sleeping racks. This liner provided the Crew a means to perform hygiene activities within a private, enclosed area that also significantly decreased the potential to grow mold. This paper will describe the development of the TeSS Hygiene Liner, its impacts on the ISS and Crew, as well as its contribution to hygiene activities used in space today.

Reid, Ethan A.↗
NASA NTRS2011

Columbus Payloads Flow Rate Anomalies

The Columbus Active Thermal Control System (ATCS) is the main thermal bus for the pressurized racks working inside the European laboratory. One of the ATCS goals is to provide proper water flow rate to each payload (P/L) by controlling actively the pressure drop across the common plenum distribution piping. Overall flow measurement performed by the Water Pump Assembly (WPA) is the only flow rate monitor available at system level and is not part of the feedback control system. At rack activation the flow rate provided by the system is derived on ground by computing the WPA flow increase. With this approach, several anomalies were raised during these 3 years on-orbit, with the indication of low flow rate conditions on the European racks FSL, BioLab, EDR and EPM. This paper reviews the system and P/Ls calibration approach, the anomalies occurred, the engineering evaluation on the measurement approach and the accuracy improvements proposed, the on-orbit test under evaluation with NASA and finally discusses possible short and long term solutions in case of anomaly confirmation.

Quaranta, Albino↗
NASA NTRS2013

CFD Lagrangian Modeling of Water Droplet Transport for ISS Hygiene Activity Application

The goal of this study was to assess the impacts of free water propagation in the Waste and Hygiene Compartment (WHC) installed in Node 3. Free water can be generated inside the WHC in small quantities due to crew hygiene activity. To mitigate potential impact of free water in Node 3 cabin the WHC doorway is enclosed by a waterproof bump-out, Kabin, with openings at the top and bottom. At the overhead side of the rack, there is a screen that prevents large drops of water from exiting. However, as the avionics fan in the WHC causes airflow toward the deck side of the rack, small quantities of free water may exit at the bottom of the Kabin. A Computational Fluid Dynamics (CFD) analysis of Node 3 cabin airflow enable identifying the paths of water transport. To simulate the droplet transport the Lagrangian discrete phase approach was used. Various initial droplet distributions were considered in the study. The droplet diameter was varied in the range of 5-20 mm. The results of the computations showed that most of the drops fall to the rack surface not far from the WHC curtain.

Son, Chang H.↗
NASA NTRS2014

Optical Multi-Gas Monitor Technology Demonstration on the International Space Station

There are a variety of both portable and fixed gas monitors onboard the International Space Station (ISS). Devices range from rack-mounted mass spectrometers to hand-held electrochemical sensors. An optical Multi-Gas Monitor has been developed as an ISS Technology Demonstration to evaluate long-term continuous measurement of 4 gases. Based on tunable diode laser spectroscopy, this technology offers unprecedented selectivity, concentration range, precision, and calibration stability. The monitor utilizes the combination of high performance laser absorption spectroscopy with a rugged optical path length enhancement cell that is nearly impossible to misalign. The enhancement cell serves simultaneously as the measurement sampling cell for multiple laser channels operating within a common measurement volume. Four laser diode based detection channels allow quantitative determination of ISS cabin concentrations of water vapor (humidity), carbon dioxide, ammonia and oxygen. Each channel utilizes a separate vertical cavity surface emitting laser (VCSEL) at a different wavelength. In addition to measuring major air constituents in their relevant ranges, the multiple gas monitor provides real time quantitative gaseous ammonia measurements between 5 and 20,000 parts-per-million (ppm). A small ventilation fan draws air with no pumps or valves into the enclosure in which analysis occurs. Power draw is only about 3 W from USB sources when installed in Nanoracks or when connected to 28V source from any EXPRESS rack interface. Internal battery power can run the sensor for over 20 hours during portable operation. The sensor is controlled digitally with an FPGA/microcontroller architecture that stores data internally while displaying running average measurements on an LCD screen and interfacing with the rack or laptop via USB. Design, construction and certification of the Multi-Gas Monitor were a joint effort between Vista Photonics, Nanoracks and NASA-Johnson Space Center (JSC). Vista Photonics developed the core technology and built the sensor. Nanoracks designed, constructed the enclosure, interfaces, and battery power management circuitry, integrated all subsystems into the enclosure, and then managed the certification tests, documentation and manifesting. The unit was calibrated in the JSC Toxicology Laboratory. The Multi-Gas Monitor is manifested to fly as a technology demonstration to the ISS in November 2013 and will operate for at least 6 months with data sent to the ground for evaluation. The primary goal is to demonstrate long term interference free operation in the real spacecraft environment.

Pilgrim, Jeffrey S.↗
NASA NTRS2015

Exercise Equipment Usability Assessment for a Deep Space Concept Vehicle

With international aspirations to send astronauts to deep space, the world is now faced with the complex problem of keeping astronauts healthy in unexplored hostile environments for durations of time never before attempted by humans. The great physical demands imparted by space exploration compound the problem of astronaut health, as the astronauts must not only be healthy, but physically fit upon destination arrival in order to perform the scientific tasks required of them. Additionally, future deep space exploration necessitates the development of environments conducive to long-duration habitation that would supplement propulsive vehicles. Space Launch System (SLS) core stage barrel sections present large volumes of robust structure that can be recycled and used for long duration habitation. This assessment will focus on one such conceptual craft, referred to as the SLS Derived Habitat (SLS-DH). Marshall Space Flight Center's (MSFC) Advanced Concepts Office (ACO) has formulated a high-level layout of this SLS-DH with parameters such as floor number and orientation, floor designations, grid dimensions, wall placement, etc. Yet to be determined, however, is the layout of the exercise area. Currently the SLS-DH features three floors laid out longitudinally, leaving 2m of height between the floor and ceilings. This short distance between levels introduces challenges for proper placement of exercise equipment such as treadmills and stationary bicycles, as the dynamic envelope for the 95th percentile male astronauts is greater than 2m. This study aims to assess the optimal equipment layout and sizing for the exercise area of this habitat. Figure 1 illustrates the layout of the DSH concept demonstrator located at MSFC. The exercise area is located on the lower level, seen here as the front half of the level occupied by a crew member. This small volume does not allow for numerous or bulky exercise machines, so the conceptual equipment has been limited to a treadmill and stationary bicycle. With the most current treadmill aboard the International Space Station (ISS), the Combined Operational Load-Bearing External Resistance Treadmill (COLBERT), being located in an International Standard Payload Rack (ISPR), the bottom of the conceptual treadmill features a height of 38in. Making the treadmill flush with the floor would be impossible in this rack configuration, as the distance from the outer wall of the spacecraft to the bottom floor would be too shallow. From preliminary sizing, the 38in required for the bottom of the treadmill combined with a 78in operational envelope for a 95th percentile may not be accommodated in the exercise area in a vertical orientation. Figure 2 demonstrates the volume required (in maroon) for an ISPR-bound treadmill in the concept demonstrator. Early indications as seen in this figure indicate that the crew members would contact the ceiling in such an arrangement. An assessment will be conducted to evaluate various orientations of exercise equipment in the concept demonstrator. Orientations to be tested include putting the bottom of the treadmill on the wall, having the treadmill at an angle in the floor both horizontally and vertically, and having a shorter (non-rack bound) treadmill in a vertical orientation on the floor. This assessment will yield findings regarding sizing of the area and how well participants feel they could exercise in such an environment. Due to the restrictions of assessing a microgravity vehicle in a normal-gravity environment, simulations in MSFC's Virtual Environments Lab (VEL) may be necessary. Final deliverables will include recommendations regarding the location and size of possible exercise equipment aboard the SLS-Derived DSH.

Rhodes, Brooke M.↗
NASA NTRS2018

Finite Element Simulations of Two Vertical Drop Tests of F-28 Fuselage Sections

In March 2017, a vertical drop test of a forward fuselage section of a Fokker F-28 MK4000 aircraft was conducted as part of a joint NASA/FAA project to investigate the performance of transport aircraft under realistic crash conditions. In June 2017, a vertical drop test was conducted of a wing-box fuselage section of the same aircraft. Both sections were configured with two rows of aircraft seats, in a triple-double configuration. A total of ten Anthropomorphic Test Devices (ATDs) were secured in seats using standard lap belt restraints. The forward fuselage section was also configured with luggage in the cargo hold. Both sections were outfitted with two hat racks, each with added ballast mass. The drop tests were performed at the Landing and Impact Research facility located at NASA Langley Research Center in Hampton, Virginia. The measured impact velocity for the forward fuselage section was 346.8-in/s onto soil. The wing-box section was dropped with a downward facing pitch angle onto a sloping soil surface in order to create an induced forward acceleration in the airframe. The vertical impact velocity of the wing-box section was 349.2-in/s. A second objective of this project was to assess the capabilities of finite element simulations to predict the test responses. Finite element models of both fuselage sections were developed for execution in LS-DYNA(Registered Trademark), a commercial explicit nonlinear transient dynamic code. The models contained accurate representations of the airframe structure, the hat racks and hat rack masses, the floor and seat tracks, the luggage in the cargo hold for the forward section, and the detailed under-floor structure in the wing-box section. Initially, concentrated masses were used to represent the inertial properties of the seats, restraints, and ATD occupants. However, later simulations were performed that included finite element representations of the seats, restraints, and ATD occupants. These models were developed to more accurately replicate the seat loading of the floor and to enable prediction of occupant impact responses. Models were executed to generate analytical predictions of airframe responses, which were compared with test data to validate the model. Comparisons of predicted and experimental structural deformation and failures were made. Finally, predicted and experimental soil deformation and crater depths were also compared for both drop test configurations.

Jackson, Karen E.↗
NASA NTRS2018

Vertical Drop Test and Simulation of a Fokker F-28 Fuselage Section

In March 2017, a vertical drop test of a 3.048-m(10-ft) section of a Fokker F-28 aircraft was conducted as a part of a joint NASA/FAA effort to investigate the performance of transport aircraft under realistic crash conditions. The section was configured with two rows of aircraft seats, in a triple-double configuration. A total of ten Anthropomorphic Test Devices (ATDs) were secured in the seats using standard seat belt restraints. The section was also configured with luggage in the cargo hold. Two hat racks were added, each with mass loading of 37.2-kg per linear meter (25-lb/ft). The drop test was performed at the Landing and Impact Research facility located at NASA Langley Research Center in Hampton, Virginia. The planned impact velocity was 9.144-m/s (360-in/s) onto soil. A second objective was to assess the capabilities of finite element simulations to predict the test response. A finite element model was developed for execution in LS-DYNA, a commercial explicit nonlinear transient dynamic code. The model contained accurate representations of the airframe structure, the hat racks and hat rack masses, the floor and seat tracks, and the luggage in the cargo hold. Concentrated masses were used to represent the inertial properties of the seats, restraints, and ATD occupants. The model was executed to generate analytical predictions of airframe responses, which were compared with test data to validate the model.

Jackson, K. E.↗
DOE OSTI2020

Solar panel field array support system and apparatus and method for construction use

A combined PV panel assembly jig and forklift transport pallet is used to assemble PV panels for transport to a field array from a protected manufacturing environment. The panels are assembled to have adhesively applied rails for transport by a robotic drone on a ground-support rack and are pre-wired. The PV panel assembly jig holds, protects, and aligns the PV panels in an upside down position, opposite to their operational position, for ease of wiring in order to decrease the manual labor required in the field. Once the pallet is transported to the load station at the end of a row of solar panel racks in the field array, a robotic loader lifts the upside down PV panels from the combined PV panel assembly jig and forklift transport pallet in an arcing overhead motion that lifts, tilts, and deposits the PV panels in an upright position at the loading station of a railed rack support as ground-mounted in a solar panel field array.

14 SOLAR ENERGY↗
DOE OSTI2016

Quantification of reactivity margin in BWR spent fuel pools

Boiling water reactor (BWR) spent fuel pool (SFP) criticality safety analyses typically use the in-core k{sub inf} criterion methodology to demonstrate margin to safety. This method models all the fuel in the SFP with a limiting lattice design at a peak reactivity just above a specific in-core k{sub inf} limit. To bound all the fuel in the pool and future fuel with reasonable margin, these limits are often set conservatively high compared to the peak reactivities of actual BWR bundle designs. When current regulatory expectations are met for the quantification of bias and uncertainty adders to the criticality analyses, some results are quantified that are close to the 0.95 in-rack reactivity limit required by 10CFR50.68. The small margins being calculated in the criticality safety analyses have resulted in concerns that SFPs have less margin to safety than was present in the past; however, inherent in the peak reactivity in-core k{sub inf} criterion methodology used to assess these margins is a very conservative model of the fuel. While it is known that all the fuel in a BWR SFP is typically burned well past its peak reactivity exposure, it has been unknown how much reactivity margin could be quantified if bundles were modeled more realistically by accounting for their actual discharge exposure profiles, operational void fractions, control histories, and depletion fuel temperatures. In this study, fuel bundles were depleted to their explicit nodal discharge exposure while accounting for actual operational conditions defined by GEH/GNF's core simulator code, PANACEA. The use of these depleted isotopics in the criticality safety analysis is commonly referred to as taking 'burn-up credit' in the analysis. This paper provides a quantification of the reactivity margin available from burn-up credit in two currently operating BWR spent fuel pools. The evaluation demonstrates that many of the bundles in these storage configurations are significantly below the in-rack reactivity limit, even with the necessary bias and uncertainty adders included. The utilization of BWR burnup credit appears to be an attractive solution to consider when evaluating options for reduced neutron absorber credit or for quantifying additional margin to the regulatory in-rack k{sub eff} limit. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2015

Overview of zion comparative analysis project for assessment of Boral{sup R} neutron absorber material performance and monitoring in spent fuel pools - 14618

Neutron absorber materials are used in spent fuel pool (SFP) storage racks to control reactivity. In the United States, about half of the operating plants currently use BORAL{sup R} neutron absorber material in SFP racks. Although operational experience spanning more than 25 years has not shown any loss of the BORAL{sup R} core material, corrosion effects, including blisters and pits, have been observed and reported. To improve understanding for the behavior of BORAL{sup R} in SFPs and evaluate the performance of monitoring approaches, EPRI initiated the Zion comparative analysis project in 2014. The Zion project involves three phases. During the first phase of the project, in situ measurements were performed on selected SFP rack panels in December 2014. Panels for in situ measurements were selected after an extensive analysis of each panel in the pool to ensure that the selected panels represented the pool in terms of variation in neutron and gamma dose values. As part of the second phase of the project, the remaining 10 coupons from the Zion SFP were shipped to a Penn State University (PSU) laboratory for analysis in January 2015. During the final phase of the project, a subset of the panels for which in situ measurements were performed will be harvested from the Zion SFP, cut into samples, and analyzed at PSU. The overall scope of this project will allow comparison of in situ measurements to real panel data as well as data obtained from coupon analysis. In this paper, an overview of the Zion comparative analysis project, the criteria and procedure for panel selection for in situ measurements and harvesting, preliminary coupon analysis results and in situ measurement results are presented. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Idaho National Laboratory’s Mobile Hot Cell Transportation: Engineering Solutions for Global Disused Sealed Radioactive Sources.

Title: Idaho National Laboratory’s Mobile Hot Cell Transportation: Engineering Solutions for Global Disused Sealed Radioactive Sources. Abstract: The Mobile Hot Cell (MHC), currently under development by Idaho National Laboratory (INL) for the Off-Site Source Recovery Project (OSRP), is designed to help international partners meet the unique challenges of end-of-life radioactive material management. The MHC will provide a critical resource for countries that require assistance securing and disposing of Disused Sealed Radioactive Sources (DSRS) and orphaned sources in challenging environments, allowing these sources to be secured against misuse and nefarious activities. The MHC is a rapidly deployable system for conditioning and preparing end-of-life radioactive sources for transportation or storage. It is designed to handle sources of up to 1,000 Ci Co-60 equivalent while maintaining full radiological and biological containment. It will be deployable within 48 hours of an alert, making it ideal for emergency situations. The MHC features an operational suite for control, support racks for electronics, pneumatics, and welding systems, and a modular robust steel structure providing radiological shielding and internal robotic support. This design allows configurations for multiple device types to be conditioned and the ability to safely manage routine issues such as leaking or damaged sources. The MHC has been designed with the transportation challenges of rapid deployment to difficult environments in mind. The system weighs approximately 150,000 pounds, with individual systems breaking down into pieces not exceeding 20,000 pounds. Components are to be transportable on standard 20ft ISO containers, with shielding shells on 20ft flat racks. It is estimated that a total of eight containers and flat racks will be required. The use of 20ft containers, as opposed to 40ft containers, minimizes the impact on less developed road infrastructures, enabling the MHC to be positioned in constrained environments such as hospital parking lots. The system’s modularity also allows for deployment using smaller equipment, such as a 10-ton boom truck or forklift, which is crucial given the potential logistical challenges in different countries. This transportation strategy, evaluated in collaboration with Utah State University, ensures the MHC can be deployed via ground, rail, sea, or air, addressing the primary concern of international transport logistics.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2024

Evaluating the potential of disaggregated memory systems for HPC applications

Summary Disaggregated memory is a promising approach that addresses the limitations of traditional memory architectures by enabling memory to be decoupled from compute nodes and shared across a data center. Cloud platforms have deployed such systems to improve overall system memory utilization, but performance can vary across workloads. High‐performance computing (HPC) is crucial in scientific and engineering applications, where HPC machines also face the issue of underutilized memory. As a result, improving system memory utilization while understanding workload performance is essential for HPC operators. Therefore, learning the potential of a disaggregated memory system before deployment is a critical step. This paper proposes a methodology for exploring the design space of a disaggregated memory system. It incorporates key metrics that affect performance on disaggregated memory systems: memory capacity, local and remote memory access ratio, injection bandwidth, and bisection bandwidth, providing an intuitive approach to guide machine configurations based on technology trends and workload characteristics. We apply our methodology to analyze thirteen diverse workloads, including AI training, data analysis, genomics, protein, fusion, atomic nuclei, and traditional HPC bookends. Our methodology demonstrates the ability to comprehend the potential and pitfalls of a disaggregated memory system and provides motivation for machine configurations. Our results show that eleven of our thirteen applications can leverage injection bandwidth disaggregated memory without affecting performance, while one pays a rack bisection bandwidth penalty and two pay the system‐wide bisection bandwidth penalty. In addition, we also show that intra‐rack memory disaggregation would meet the application's memory requirement and provide enough remote memory bandwidth.

Ding, Nan↗
DOE OSTI2018

Efficiency and Reliability Analyses of AC and 380V DC Distribution in Data Centers

Modern data centers consume large amounts of electricity, resulting in high operational costs. The efficiency of a data center power distribution system can be increased, and the operational cost reduced, if the number of power conversion stages can be minimized and more efficient converters, such as wide bandgap (WBG) converters, are used. This can be achieved by using DC distribution voltage at the rack level by eliminating extra conversion stages. In this paper, benchmarks for both AC and 380V DC data centers were developed and efficiency analyses were performed for an entire year. The impact of integrating photovoltaic (PV) systems into the data centers has also been analyzed in both cases in terms of efficiency. The results show that 380V DC data centers are more efficient than AC data centers with and without PV integration. Furthermore, the reliability of an AC system was compared to a 380V DC architecture with Tier-IV standard. Monte-Carlo simulations were used to perform reliability analyses for different levels of redundancy in the Uninterruptible Power Supply (UPS) system for both cases. In conclusion, the simulation results showed that the 380V DC distribution system had a higher level of reliability than the AC distribution system in data centers.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2016

Analysis of Irradiance Models for Bifacial PV Modules

We describe and compare two methods for modeling irradiance on the back surface of rack-mounted bifacial PV modules: view factor models and ray-tracing simulations. For each method we formulate one or more models and compare each model with irradiance measurements and short circuit current for a bifacial module mounted a fixed tilt rack with three other similarly sized modules. Our analysis illustrates the computational requirements of the different methods and provides insight into their practical applications. We find a level of consistency among the models which indicates that consistent models may be obtained by parameter calibrations.

14 SOLAR ENERGY↗
DOE OSTI2013

Feng shui of supercomputer memory: positional effects in DRAM and SRAM faults. In: SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

Several recent publications confirm that faults are common in high-performance computing systems. Therefore, further attention to the faults experienced by such computing systems is warranted. In this paper, we present a study of DRAM and SRAM faults in large high-performance computing systems. Our goal is to understand the factors that influence faults in production settings. We examine the impact of aging on DRAM, finding a marked shift from permanent to transient faults in the first two years of DRAM lifetime. We examine the impact of DRAM vendor, finding that fault rates vary by more than 4x among vendors. We examine the physical location of faults in a DRAM device and in a data center; contrary to prior studies, we find no correlations with either. Finally, we study the impact of altitude and rack placement on SRAM faults, finding that, as expected, altitude has a substantial impact on SRAM faults, and that top of rack placement correlates with 20% higher fault rate.

Sridharan, Vilas↗