Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Data Transfer”

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 559 records · Page 31
All sourcesNASA NTRSDOE OSTI
NASA NTRS2003

Reusable Rack Interface Controller Common Software for Various Science Research Racks on the International Space Station

The purpose of the EXPRESS (Expedite the PRocessing of Experiments to Space Station) rack project is to provide a set of predefined interfaces for scientific payloads which allow rapid integration into a payload rack on International Space Station (ISS). VxWorks' was selected as the operating system for the rack and payload resource controller, primarily based on the proliferation of VME (Versa Module Eurocard) products. These products provide needed flexibility for future hardware upgrades to meet everchanging science research rack configuration requirements. On the International Space Station, there are multiple science research rack configurations, including: 1) Human Research Facility (HRF); 2) EXPRESS ARIS (Active Rack Isolation System); 3) WORF (Window Observational Research Facility); and 4) HHR (Habitat Holding Rack). The RIC (Rack Interface Controller) connects payloads to the ISS bus architecture for data transfer between the payload and ground control. The RIC is a general purpose embedded computer which supports multiple communication protocols, including fiber optic communication buses, Ethernet buses, EIA-422, Mil-Std-1553 buses, SMPTE (Society Motion Picture Television Engineers)-170M video, and audio interfaces to payloads and the ISS. As a cost saving and software reliability strategy, the Boeing Payload Software Organization developed reusable common software where appropriate. These reusable modules included a set of low-level driver software interfaces to 1553B. RS232, RS422, Ethernet buses, HRDL (High Rate Data Link), video switch functionality, telemetry processing, and executive software hosted on the FUC computer. These drivers formed the basis for software development of the HRF, EXPRESS, EXPRESS ARIS, WORF, and HHR RIC executable modules. The reusable RIC common software has provided extensive benefits, including: 1) Significant reduction in development flow time; 2) Minimal rework and maintenance; 3) Improved reliability; and 4) Overall reduction in software life cycle cost. Due to the limited number of crew hours available on ISS for science research, operational efficiency is a critical customer concern. The current method of upgrading RIC software is a time consuming process; thus, an improved methodology for uploading RIC software is currently under evaluation.

Lu, George C.↗
DOE OSTI2020

Robust Online Sequential RVFLNs for Data Modeling of Dynamic Time-Varying Systems with Application of an Ironmaking Blast Furnace

In a world where the increasing complexity of modern industrial processes brings difficulties for accurate mathematical modeling, taking advantage of data has become an efficient solution to complex dynamic process modeling issue. In this paper, we develop a novel robust online sequential version of random vector functional-link networks (RVFLNs) for data-driven modeling of dynamic time-varying system and applied it in a blast furnace (BF) ironmaking process. First, to overcome the time-varying dynamics of process and to enable the RVFLNs to learn online with avoiding data saturation, an improved online sequential version of RVFLNs (OS-RFVLNs) is first presented by online sequential learning with forgetting factor. This improved OS-RVFLNs algorithm is not only suitable for the real-time and large data transfer situation, but also can adjust the sensitivity of the algorithm to different samples with the help of the introduced forgetting factor. Second, since the output weights of the improved OS-RVFLNs as well as other RVFLNs algorithms are obtained by the least squares approach, a robustness problem may occur when the training dataset is contaminated with various outliers. To solve this problem, a Cauchy distribution weighted M-estimator is introduced to improve the robustness of the improved OS- RVFLNs. For this proposed robust OS-RVFLNs (R-OS- RVFLNs), since the weights of different outlier data are properly determined by the Cauchy distribution function, their corresponding contribution on modeling can be properly distinguished. Thus robust and better modeling results can be achieved. Experiments using actual industrial data of BF ironmaking process and comparative studies have demonstrated that the proposed method produces a better estimation accuracy and stronger robustness than other methods.

Blast furnace (BF), Modelling, Dynamic systems↗
NASA NTRS1999

ACTS 118x: High Speed TCP Interoperability Testing

With the recent explosion of the Internet and the enormous business opportunities available to communication system providers, great interest has developed in improving the efficiency of data transfer over satellite links using the Transmission Control Protocol (TCP) of the Internet Protocol (IP) suite. The NASA's ACTS experiments program initiated a series of TCP experiments to demonstrate scalability of TCP/IP and determine to what extent the protocol can be optimized over a 622 Mbps satellite link. Through partnerships with the government technology oriented labs, computer, telecommunication, and satellite industries NASA Glenn was able to: (1) promote the development of interoperable, high-performance TCP/IP implementations across multiple computing / operating platforms; (2) work with the satellite industry to answer outstanding questions regarding the use of standard protocols (TCP/IP and ATM) for the delivery of advanced data services, and for use in spacecraft architectures; and (3) conduct a series of TCP/IP interoperability tests over OC12 ATM over a satellite network in a multi-vendor environment using ACTS. The experiments' various network configurations and the results are presented.

Brooks, David E.↗
NASA NTRS2010

Implementation of a Relay Coordination System for the Mars Network

Mars network relay operations involve the coordination of lander and orbiter teams through long-term and short-term planning, tactical changes and post-pass analysis. Much of this coordination is managed through email traffic and point-to-point file data exchanges. It is often difficult to construct a complete and accurate picture of the relay situation at any given moment, as there is no centralized store of correlated relay data. The Mars Relay Operations Service (MaROS) is being implemented to address the problem of relay coordination for current and next-generation relay missions. The service is provided for the purpose of coordinating communications sessions between landed spacecraft assets and orbiting spacecraft assets at Mars. The service centralizes a set of functions previously distributed across multiple spacecraft operations teams, and as such greatly improves visibility into the end-to-end strategic coordination process. Most of the process revolves around the scheduling of communications sessions between the spacecraft during periods of time when a landed asset on Mars is geometrically visible by an orbiting spacecraft. These "relay" sessions are used to transfer data both to and from the landed asset via the orbiting asset on behalf of Earth-based spacecraft operators. This paper will discuss the relay coordination problem space, overview the architecture and design selected to meet system requirements, and describe the first phase of system implementation

Allard, Daniel A.↗
NASA NTRS1991

Thermal performance of a liquid hydrogen tank multilayer insulation system at warm boundary temperatures of 630, 530, and 152 R

The results are presented of a study conducted to obtain experimental heat transfer data on a liquid hydrogen tank insulated with 34 layers of MLI (multilayer insulation) for warm side boundary temperatures of 630, 530, and 150 R. The MLI system consisted of two blankets, each blanket made up of alternate layers of double silk net (16 layers) and double aluminized Mylar radiation shields (15 layers) contained between two cover sheets of Dacron scrim reinforced Mylar. The insulation system was designed for and installed on a 87.6 in diameter liquid hydrogen tank. Nominal layer density of the insulation blankets is 45 layers/in. The insulation system contained penetrations for structural support, plumbing, and electrical wiring that would be representative of a cryogenic spacecraft. The total steady state heat transfer rates into the test tank for shroud temperatures of 630, 530, 152 R were 164.4, 95.8, and 15.9 BTU/hr respectively. The noninsulation heat leaks into the tank (12 fiberglass support struts, tank plumbing, and instrumentation lines) represent between 13 to 17 pct. of the total heat input. The heat input values would translate to liquid H2 losses of 2.3, 1.3, and 0.2 pct/day, with the tank held at atmospheric pressure.

Stochl, Robert J.↗
NASA NTRS1991

Thermal performance of a liquid hydrogen tank multilayer insulation system at warm boundary temperatures of 630, 530, and 152 R

The results are presented of a study conducted to obtain experimental heat transfer data on a liquid hydrogen tank insulated with 34 layers of MLI (multilayer insulation) for warm side boundary temperatures of 630, 530, and 150 R. The MLI system consisted of two blankets, each blanket made up of alternate layers of double silk net (16 layers) and double aluminized Mylar radiation shields (15 layers) contained between two cover sheets of Dacron scrim reinforced Mylar. The insulation system was designed for and installed on an 87.6 in. diameter liquid hydrogen tank. Nominal layer density of the insulation blankets is 45 layers/in. The insulation system contained penetrations for structural support, plumbing, and electrical wiring that would be representative of a cryogenic spacecraft. The total steady state heat transfer rates into the test tank for shroud temperatures of 630, 530, 152 R were 164.4, 95.8, and 15.9 BTU/hr, respectively. The noninsulation heat leaks into the tank (12 fiberglass support struts, tank plumbing, and instrumentation lines) represent between 13 to 17 pct. of the total heat input. The heat input values would translate to liquid H2 losses of 2.3, 1.3, and 0.2 pct/day, with the tank held at atmospheric pressure.

Stochl, Robert J.↗
NASA NTRS1984

Liner environment effects study

The Liner Environment Effects Study Program is aimed at establishing a broad heat transfer data base under controlled experimental conditions by quantifying the effects of the combustion system conditions on the combustor liner thermal loading and on the flame radiation characteristics. Five liner concepts spanning the spectrum of liner design technology from the very simple to the most advanced concepts are investigated. These concepts comprise an uncooled liner, a conventional film cooled liner, an impingement/film cooled liner, a laser drilled liner approaching the concept of a porous wall, and a siliconized silicon carbide ceramic liner. Effect of fuel type is covered by using fuels containing 11.8, 12.8, and 14% hydrogen. Tests at 100, 200, and 300 psia provide a basis for evaluating the effect of pressure on the heat transfer. The effects of the atomization quality and spray characteristics are examined by varying the fuel spray Sauter mean diameter and the spray angle. Additional varied parameters include reference velocity, a wide range of equivalence ratio, cooling flow rate, coolant temperature and the velocity of the coolant stream on the backside of the liner.

Venkataramani, K. S.↗
NASA NTRS2021

Hypersonic Second-Mode Instability Response to Shaped Roughness

An experimental campaign was conducted on a 7-degree half-angle cone in the NASA Langley Research Center 20-Inch Mach 6 Wind Tunnel to examine the influence of arrays of regularly spaced roughness elements on instability growth and transition. The primary element shape was a pair of elliptical planform ramps that were inclined at equal and opposite angles with respect to the local streamwise direction. The element shapes were designed to induce transient growth disturbances that would lead to sustained azimuthal modulation of the boundary layer flow while limiting the nearfield disturbances to avoid an immediate, i.e., effective tripping of the boundary layer. The bulk of the run matrix consisted of testing different element height sat free stream unit Reynolds numbers ranging from 9.8 to 13.1 million per meter. Other element shapes previously designed for tripping hypersonic boundary layers were also implemented. The model was instrumented with surface mounted Kulite ® and PCB ® pressure transducers and thermocouples. Spectra from the PCBs ® indicated clear suppression of the second-mode instability; however, neither the PCB ® spectra nor the heat transfer data presented strong evidence for delayed turbulent flow. Complementary stability computations likewise demonstrated second-mode reduction, particularly just downstream of the roughness, but also revealed a rise in first mode (streak-instability) amplitudes from the baseline that was likely responsible for the earlier transition observed for taller roughness cases.

Hypersonics↗
NASA NTRS2021

Hypersonic Second-Mode Instability Response to Shaped Roughness

An experimental campaign was conducted on a 7-degree half-angle cone in the NASA Langley Research Center 20-Inch Mach 6 Wind Tunnel to examine the influence of arrays of regularly spaced roughness elements on instability growth and transition. The primary element shape was a pair of elliptical planform ramps that were inclined at equal and opposite angles with respect to the local streamwise direction. The element shapes were designed to induce transient growth disturbances that would lead to sustained azimuthal modulation of the boundary layer flow while limiting the nearfield disturbances to avoid an immediate, i.e., effective tripping of the boundary layer. The bulk of the run matrix consisted of testing different element height sat free stream unit Reynolds numbers ranging from 9.8 to 13.1 million per meter. Other element shapes previously designed for tripping hypersonic boundary layers were also implemented. The model was instrumented with surface mounted Kulite ® and PCB ® pressure transducers and thermocouples. Spectra from the PCBs ® indicated clear suppression of the second-mode instability; however, neither the PCB ® spectra nor the heat transfer data presented strong evidence for delayed turbulent flow. Complementary stability computations likewise demonstrated second-mode reduction, particularly just downstream of the roughness, but also revealed a rise in first mode (streak-instability) amplitudes from the baseline that was likely responsible for the earlier transition observed for taller roughness cases.

Hypersonics↗
NASA NTRS2021

Hypersonic Second Mode Instability Response to Shaped Roughness

An experimental campaign was conducted on a 7-degree half-angle cone in the NASA Langley Research Center 20-Inch Mach 6 Wind Tunnel to examine the influence of arrays of regularly spaced roughness elements on instability growth and transition. The primary element shape was a pair of elliptical planform ramps that were inclined at equal and opposite angles with respect to the local streamwise direction. The element shapes were designed to induce transient growth disturbances that would lead to sustained azimuthal modulation of the boundary layer flow while limiting the nearfield disturbances to avoid an immediate, i.e., effective tripping of the boundary layer. The bulk of the run matrix consisted of testing different element height sat free stream unit Reynolds numbers ranging from 9.8 to 13.1 million per meter. Other element shapes previously designed for tripping hypersonic boundary layers were also implemented. The model was instrumented with surface mounted Kulite ® and PCB ® pressure transducers and thermocouples. Spectra from the PCBs ® indicated clear suppression of the second-mode instability; however, neither the PCB ® spectra nor the heat transfer data presented strong evidence for delayed turbulent flow. Complementary stability computations likewise demonstrated second-mode reduction, particularly just downstream of the roughness, but also revealed a rise in first mode (streak-instability) amplitudes from the baseline that was likely responsible for the earlier transition observed for taller roughness cases.

Hypersonics↗
NASA NTRS2011

SpaceCube 2.0: An Advanced Hybrid Onboard Data Processor

The SpaceCube 2.0 is a compact, high performance, low-power onboard processing system that takes advantage of cutting-edge hybrid (CPU/FPGA/DSP) processing elements. The SpaceCube 2.0 design concept includes two commercial Virtex-5 field-programmable gate array (FPGA) parts protected by gradiation hardened by software" technology, and possesses exceptional size, weight, and power characteristics [5x5x7 in., 3.5 lb (approximately equal to 12.7 x 12.7 x 17.8 cm, 1.6 kg) 5-25 W, depending on the application fs required clock rate]. The two Virtex-5 FPGA parts are implemented in a unique back-toback configuration to maximize data transfer and computing performance. Draft computing power specifications for the SpaceCube 2.0 unit include four PowerPC 440s (1100 DMIPS each), 500+ DSP48Es (2x580 GMACS), 100+ LVDS high-speed serial I/Os (1.25 Gbps each), and 2x190 GFLOPS single-precision (65 GFLOPS double-precision) floating point performance. The SpaceCube 2.0 includes PROM memory for CPU boot, health and safety, and basic command and telemetry functionality; RAM memory for program execution; and FLASH/EEPROM memory to store algorithms and application code for the CPU, FPGA, and DSP processing elements. Program execution can be reconfigured in real time and algorithms can be updated, modified, and/or replaced at any point during the mission. Gigabit Ethernet, Spacewire, SATA and highspeed LVDS serial/parallel I/O channels are available for instrument/sensor data ingest, and mission-unique instrument interfaces can be accommodated using a compact PCI (cPCI) expansion card interface. The SpaceCube 2.0 can be utilized in NASA Earth Science, Helio/Astrophysics and Exploration missions, and Department of Defense satellites for onboard data processing. It can also be used in commercial communication and mapping satellites.

Lin, Michael↗
NASA NTRS2018

Time and Laser Ranging: A Window of Opportunity for Geodesy, Navigation, and Meteorology

Recent progress in the domain of time and frequency (T/F) standards requires important improvements in existing timedistribution links, in term of accuracy in particular. Satellite Laser Ranging (SLR) has proven to be a fundamental tool,offering a straightforward, conceptually simple, highly accurate, and unambiguous observable. Several time transfers by laserlink projects have been carried out over the past 10 years with numerous scientific and metrological objectives. Depending on the mission, SLR is used to transmit time over two-way or one-way distances from 500 to several millions of kilometers.The following missions and their objectives employed this technique: European Laser Timing (ELT, expected in 2020) at 450 km, Time Transfer by Laser Link (T2L2) at 1336 km, Laser Time Transfer at 36,000 km, Lunar Reconnaissance Orbiter at 350,000 km, and MErcury Surface, Space ENvironment, GEochemistry, and Ranging at tens of million km. This article describes the synergy between SLR and T/F technologies developed on the ground and in space and as well as the state of the art of their exploitation. The performance and sources of limitation of such space missions are analyzed. It shows that currentand future challenges lie in the improvement in the time accuracy and stability of the time for ground geodetic observatories.The role of the next generation of SLR systems is emphasized both in space and at ground level, from the point of view of Global Geodetic Observing System and valuable exploitation of the synergy between time synchronization, ranging, and data transfer.

Laser ranging↗
NASA NTRS2018

Dual-mode Propulsion System Enabling Cubesat Exploration of the Solar System

It is apparent the cost of planetary exploration is rising as mission budgets are declining. Currently small scientific beds geared to performing limited tasks are being developed and launched into low earth orbit (LEO) in the form of small-scale satellite units, i.e., CubeSats. These micro- and nano-satellites are gaining popularity among the university and science communities due to their relatively low cost and design flexibility. To date these small units have been limited to performing tasks in LEO utilizing solar-based power. If a reasonable propulsion system could be developed, these CubeSat platforms could perform exploration of various extra-terrestrial bodies within the solar system engaging a broader range of researchers. Additionally, being mindful of mass, smaller cheaper launch vehicles (~1,000 kg to LEO) can be targeted. This, in effect, allows for beneficial exploration to be conducted within limited budgets.Researchers at the Center for Space Nuclear Research (CSNR) are proposing a low mass, radioisotope-based, dual-mode propulsion system capable of extending the exploration realm of these CubeSats out of LEO.The proposed radioisotope-based system would leverage the high specific energies [J/kg] associated with radioisotope materials and enhance their inherent low specific powers [W/g]. This is accomplished by accumulating thermal energy from nuclear decay within a central core over time. This allows for significant amounts of power to be transferred to a flowing gas over short periods of time. In the proposed configuration the stored energy can be utilized in two ways: (1) with direct propellant injection to the core, the energy can be converted into thrust through the use of a converging-diverging nozzle and (2) by flowing a working fluid through the core and subsequent Brayton engine, energy within the core can be converted to electrical energy. The first scenario achieves moderate ranges of thrust, but at a higher Isp than traditional chemical-based systems. The second scenario allows for the production of electrical power, which is then available for electric-based propulsion. Additionally, once at location the production of electrical power can be dedicated to the payload's communication system for data transfer. Ultimately, the proposed dual-mode propulsion platform capitalizes on the benefits of two types of propulsion methods the thrust of thermal propulsion ideal for quick orbital maneuvers and the specific impulse of electric propulsion ideal for efficient interplanetary travel. Overall, the system is functioning as a radioisotope thermal rocket (RTR).In this study the RTR concept is being developed as an in-space propulsion system to deliver a 6U CubeSat payload to the orbit of the Saturnian moon - Enceladus. Additionally, this study will develop an entire mission architecture for Enceladus targeting a total allowable launch mass of 1,000 kg.

Cubesats↗
NASA NTRS2018

Time and Laser Ranging: A Window of Opportunity for Geodesy, Navigation and Metrology

Recent progress in the domain of time and frequency (T/F) standards requires important improvements of existing time distribution links. Among these, the accuracy of time transfer is actually an important part of the concerns in order to establish and maintain time & space references from ground and/or space facilities. Several time transfers by laser link projects have been carried out over the past 10 years with numerous scientific and metrological objectives. Satellite Laser ranging (SLR) has proven to be a fundamental tool, offering a straightforward, conceptually simple, highly accurate and unambiguous observable. Depending on the mission, LR is used to transmit time over two-way or one-way distances from 500 to several millions of km. The following missions and their objectives employed this technique: European Laser Timing (ELT) at 450 km, Time Transfer by Laser Link (T2L2) at 1,336 km, Laser Time Transfer (LTT) at 36,000 km, Lunar Reconnaissance Orbiter (LRO) at 350,000 km, and MErcury Surface, Space ENvironment, GEochemistry, and Ranging (MESSENGER) at tens of million km. This article describes the synergy between SLR and T/F technologies developed on the ground and in space and as well as the state of the art of their exploitation. The performance and sources of limitation of such space missions are analyzed. It shows that current and future challenges lie in the improvement of the time accuracy and stability of the time for ground geodetic observatories. The role of the next generation of SLR systems is emphasized both in space and at ground level, from the point of view of GGOS and valuable exploitation of the synergy between time synchronization, ranging and data transfer.

P. Exertier↗
NASA NTRS2000

Acceptance Testing of a Satellite SCADA Photovoltaic-Diesel Hybrid System

Satellite Supervisory Control and Data Acquisition (SCADA) of a Photovoltaic (PV)/diesel hybrid system was tested using NASA's Advanced Communication Technology Satellite (ACTS) and Ultra Small Aperture Terminal (USAT) ground stations. The setup consisted of a custom-designed PV/diesel hybrid system, located at the Florida Solar Energy Center (FSEC), which was controlled and monitored at a "remote" hub via Ka-band satellite link connecting two 1/4 Watt USATs in a SCADA arrangement. The robustness of the communications link was tested for remote monitoring of the health and performance of a PV/diesel hybrid system, and for investigating load control and battery charging strategies to maximize battery capacity and lifetime, and minimize loss of critical load probability. Baseline hardware performance test results demonstrated that continuous two-second data transfers can be accomplished under clear sky conditions with an error rate of less than 1%. The delay introduced by the satellite (1/4 sec) was transparent to synchronization of satellite modem as well as to the PV/diesel-hybrid computer. End-to-end communications link recovery times were less than 36 seconds for loss of power and less than one second for loss of link. The system recovered by resuming operation without any manual intervention, which is important since the 4 dB margin is not sufficient to prevent loss of the satellite link during moderate to heavy rain. Hybrid operations during loss of communications link continued seamlessly but real-time monitoring was interrupted. For this sub-tropical region, the estimated amount of time that the signal fade will exceed the 4 dB margin is about 10%. These results suggest that data rates of 4800 bps and a link margin of 4 dB with a 1/4 Watt transmitter are sufficient for end-to-end operation in this SCADA application.

Kalu, A.↗
NASA NTRS2006

Considerations for Isochronous Data Services over the Proximity-1 Space Link

Future mission concepts for robotic and human explorations will involve a high level of real time control/monitoring operations such as tele-operation for spacecraft rendezvous and surface mobile platforms carrying life-support equipments. The timely dissemination of voice, command, and real-time telemetry for monitoring and coordination purposes is critical for mission success. It is envisioned that future missions will require a network infrastructure capable of supporting isochronous data services. The CCSDS Proximity-1 Space Link Protocol1 could be used to provide isochronous service over the surface-to-Earth relay as well as "beyond-the-horizon" communications between distant Lunar or Mars surface elements. This paper will analyze the latency, jitter, and throughput performance of the Proximity-1 protocol for isochronous applications. In particular we will focus on constrained scenarios where the protocol operates in full-duplex mode, carrying isochronous traffic in one direction and error-controlled traffic in the other direction. We analyze the impact of the strict priority scheme in Proximity-1 on delay jitter and the impact of the isochronous traffic on the efficiency of the reliable data transfer in the other direction, and discuss methods for performance optimization. In general, jitter performance is driving by relative loading of isochronous traffic on the forward link compared to the acknowledgement traffic. Under light loading condition, the upper-bound of the delay jitter is the transmission duration of an acknowledgement frame on the forward link; for higher loading scenarios, the maximum jitter is scaled up by the inverse of the residual bandwidth, i.e., the spare capacity available in the forward link to carry isochronous traffic.

isochronous communications↗
DOE OSTI2026

Low power on-chip data transmission for wafer-scale monolithic active pixel sensors

Here, this paper details the implementation of the digital pulse shaping subsystem within the Backbone Transmission Line Encoding (BTLE) driver, a low-power, long-distance on-chip data transmission solution designed in a 65 nm CMOS process. Digital pulse shaping is critical for minimizing inter-symbol interference (ISI) caused by bandwidth limitations of on-chip interconnects, especially in wafer-scale monolithic active pixel sensors (MAPS). A duobinary encoder coupled with a parallelized polyphase finite impulse response (FIR) filter is used for efficient shaping of the transmitted signal spectrum. This reconfigurable architecture achieves reliable 160 Mb/s data transfer over a 10 cm on-chip link, as validated by simulations demonstrating low power consumption (FoM 37.3 fJ/bit/mm of transmission line length) and effective ISI mitigation.

47 OTHER INSTRUMENTATION↗
NASA NTRS1994

Characterizing parallel file-access patterns on a large-scale multiprocessor

Rapid increases in the computational speeds of multiprocessors have not been matched by corresponding performance enhancements in the I/O subsystem. To satisfy the large and growing I/O requirements of some parallel scientific applications, we need parallel file systems that can provide high-bandwidth and high-volume data transfer between the I/O subsystem and thousands of processors. Design of such high-performance parallel file systems depends on a thorough grasp of the expected workload. So far there have been no comprehensive usage studies of multiprocessor file systems. Our CHARISMA project intends to fill this void. The first results from our study involve an iPSC/860 at NASA Ames. This paper presents results from a different platform, the CM-5 at the National Center for Supercomputing Applications. The CHARISMA studies are unique because we collect information about every individual read and write request and about the entire mix of applications running on the machines. The results of our trace analysis lead to recommendations for parallel file system design. First the file system should support efficient concurrent access to many files, and I/O requests from many jobs under varying load conditions. Second, it must efficiently manage large files kept open for long periods. Third, it should expect to see small requests predominantly sequential access patterns, application-wide synchronous access, no concurrent file-sharing between jobs appreciable byte and block sharing between processes within jobs, and strong interprocess locality. Finally, the trace data suggest that node-level write caches and collective I/O request interfaces may be useful in certain environments.

Purakayastha, Apratim↗