Co-design of System Software for Compute Accelerators and SmartNICs.
Abstract not provided.
SEARCH · Engineering Papers
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.
Abstract not provided.
Commercial buildings are responsible for approximately 20 percent of the total United States energy consumption and greenhouse gas emissions. Over 85 percent of these buildings lack building automation systems. Many of these buildings are small (<50,000 square feet), underserved, and use rooftop units for heating, ventilation, and air-conditioning needs. Because these buildings lack proper energy management systems, they have several operational deficiencies that lead to excess energy consumption. Studies have shown that managing the rooftop units heating and cooling set points, schedules, setbacks, and optimal start can result in 20 to 25 percent reduction in electricity consumption in small commercial buildings. In addition, improving demand flexibility of these buildings will result additional cost savings for the building owner. Therefore, the Department of Energy’s Building Technologies Office approved a project to address the needs for small commercial buildings. The project is led by Pacific Northwest National Laboratory (PNNL) with Intellimation LLC as the cooperative research and development agreement partner. The primary goal of the project is to develop and validate an autonomous energy management software (AEMS) system that will continuously optimize small commercial building operations by minimizing energy consumption and cost, while providing a solution for maximizing decarbonization benefits from electrification of buildings. The work will leverage the vast experience of PNNL research and development staff who have over two decades of experience in developing and successfully transferring software technologies to the private sector. This solution will be jointly developed with Intellimation, a company that plans to use it to scale their building energy efficiency (EE) and grid services offering. Widespread deployment of the AEMS system will improve the EE and demand flexibility of the building commercial building stock. It should also support cities and states in meeting their climate change mitigation goals. This document describes the various EE and grid service features of the AEMS system, infrastructure and data required to implement those features, and how the features should be automated. It also details how the various features will be tested and validated, including field validation. The document also details what flexibility the users have and how they will be able to leverage those capabilities exercise those. The intent is to create an AEMS system that would support scalable deployment, requires minimal configuration, and is easy to maintain over its expected lifespan. The initial alpha release of AEMS system is planned for March 2023, and the beta release is planned for the summer of 2023. The final release is planned for March 2024. Section 2 of the report documents the relevant building types that AEMS is suitable for. Section 3 documents EE features that will be supported. It will also include the data requirements, hardware requirements, implementation details, and how EE features will be tested and validated. Grid service features will be documented in section 4, including data requirements, hardware requirements, implementation details, and how the services will be tested and validated. Planned next steps are described in section 5.
There are no author-identified significant results in this report.
The FORTRAN Static Source Code Analyzer Program (SAP) User's Guide (Revision 1) is presented. SAP is a software tool designed to assist Software Engineering Laboratory (SEL) personnel in conducting studies of FORTRAN programs. SAP scans FORTRAN source code and produces reports that present statistics and measures of statements and structures that make up a module. This document is a revision of the previous SAP user's guide, Computer Sciences Corporation document CSC/TM-78/6045. SAP Revision 1 is the result of program modifications to provide several new reports, additional complexity analysis, and recognition of all statements described in the FORTRAN 77 standard. This document provides instructions for operating SAP and contains information useful in interpreting SAP output.
RULCHK computer program aids in structuring layers of information pertaining to part or assembly designed with software described in article "Software for Drawing Design Details Concurrently" (MFS-28444). Checks and optionally updates structure of layers for part. Enables designer to construct model and annotate its documentation without burden of manually layering part to conform to standards at design time.
A patched conic trajectory optimization program MIDAS is described that was developed to investigate a wide variety of complex ballistic heliocentric transfer trajectories. MIDAS includes the capability of optimizing trajectory event times such as departure date, arrival date, and intermediate planetary flyby dates and is able to both add and delete deep space maneuvers when dictated by the optimization process. Both powered and unpowered flyby or gravity assist trajectories of intermediate bodies can be handled and capability is included to optimize trajectories having a rendezvous with an intermediate body such as for a sample return mission. Capability is included in the optimization process to constrain launch energy and launch vehicle parking orbit parameters.
The project utilizes design automation software tools to design, simulate, and fabricate a pH meter integrated circuit (IC) system including a successive approximation type seven-bit analog to digital converter circuits using a 1.25 micron N-Well CMOS MOSIS process. The input voltage ranges from 0.5 to 1.0 V derived from a special type pH sensor, and the output is a three-digit decimal number display of pH with one decimal point.
New features have been added to IODA to support testing of lightweight optics. Algorithms have been developed to predict actuator influence functions and positions to correct for gravity loading effects. New features have also been added to import measured test data and to compare measured data with predicted data.
John Thousand of Wolverine Western Corp. put his aerospace group to work on an unfamiliar job, designing a brake drum using computer design techniques. Computer design involves creation of a mathematical model of a product and analyzing its effectiveness in simulated operation. Technique enables study of performance and structural behavior of a number of different designs before settling on a final configuration. Wolverine employees attacked a traditional brake drum problem, the sudden buildup of heat during fast and repeated braking. Part of brake drum not confined tends to change its shape under combination of heat, physical pressure and rotational forces, a condition known as bellmouthing. Since bellmouthing is a major factor in braking effectiveness, a solution of problem would be a major advance in automotive engineering. A former NASA employee, now a Wolverine employee, knew of a series of NASA computer programs ideally suited to confronting bellmouthing. Originally developed as aids to rocket engine nozzle design, it's capable of analyzing problems generated in a rocket engine or automotive brake drum by heat, expansion, pressure and rotational forces. Use of these computer programs led to new brake drum concept featuring a more durable axle, and heat transfer ribs, or fins, on hub of drum.
The current program status reviewed by this viewgraph presentation includes: 1) New Evaluation Results; 2) Algorithm Improvement Investigations; 3) Electronic Hardware Design; 4) Software Hardware Interface Design.
Engineers in the Integrated Design and Analysis Division (EV30) use a variety of different tools to aid in the design and analysis of the Ares I vehicle. One primary tool in use is Pro-Engineer. Pro-Engineer is a computer-aided design (CAD) software that allows designers to create computer generated structural models of vehicle structures. For the Upper State thrust cone, Pro-Engineer was used to assist in the design of a layout for two camera housings. These cameras observe the separation between the first and second stage of the Ares I vehicle. For the Ares I-X, one standard speed camera was used. The Ares I design calls for two separate housings, three cameras, and a lighting system. With previous design concepts and verification strategies in mind, a new layout for the two camera design concept was developed with members of the EV32 team. With the new design, Pro-Engineer was used to draw the layout to observe how the two camera housings fit with the thrust cone assembly. Future analysis of the camera housing design will verify the stability and clearance of the camera with other hardware present on the thrust cone.
Source Lines Counter (SLiC) is a software utility designed to measure software source code size using logical source statements and other common measures for 22 of the programming languages commonly used at NASA and the aerospace industry. Such metrics can be used in a wide variety of applications, from parametric cost estimation to software defect analysis. SLiC has a variety of unique features such as automatic code search, automatic file detection, hierarchical directory totals, and spreadsheet-compatible output. SLiC was written for extensibility; new programming language support can be added with minimal effort in a short amount of time. SLiC runs on a variety of platforms including UNIX, Windows, and Mac OSX. Its straightforward command-line interface allows for customization and incorporation into the software build process for tracking development metrics. T
NASA's Space Launch System (SLS) is an advanced launch vehicle for a new era of exploration beyond earth's orbit (BEO). The world's most powerful rocket, SLS, will launch crews of up to four astronauts in the agency's Orion spacecraft on missions to explore multiple deep-space destinations. Boeing is developing the SLS core stage, including the avionics that will control vehicle during flight. The core stage will be built at NASA's Michoud Assembly Facility (MAF) in New Orleans, LA using state-of-the-art manufacturing equipment. At the same time, the rocket's avionics computer software is being developed here at Marshall Space Flight Center in Huntsville, AL. At Marshall, the Flight and Ground Software division provides comprehensive engineering expertise for development of flight and ground software. Within that division, the Software Systems Engineering Branch's test and verification (T&V) team uses an agile test approach in testing and verification of software. The agile software test method opens the door for regular short sprint release cycles. The idea or basic premise behind the concept of agile software development and testing is that it is iterative and developed incrementally. Agile testing has an iterative development methodology where requirements and solutions evolve through collaboration between cross-functional teams. With testing and development done incrementally, this allows for increased features and enhanced value for releases. This value can be seen throughout the T&V team processes that are documented in various work instructions within the branch. The T&V team produces procedural test results at a higher rate, resolves issues found in software with designers at an earlier stage versus at a later release, and team members gain increased knowledge of the system architecture by interfacing with designers. SLS Flight Software teams want to continue uncovering better ways of developing software in an efficient and project beneficial manner. Through agile testing, there has been increased value through individuals and interactions over processes and tools, improved customer collaboration, and improved responsiveness to changes through controlled planning. The presentation will describe agile testing methodology as taken with the SLS FSW Test and Verification team at Marshall Space Flight Center.
No abstract available
Summary of technical results of Blackcomb Memory Devices We explored various different memory technologies (STTRAM, PCRAM, FeRAM, and ReRAM). The progress can be classified into three categories, below. Modeling and Tool Releases Various modeling tools have been developed over the last decade to help in the design of SRAM or DRAM-based memory hierarchies. To explore new design opportunities that NVM technologies can bring to the designers, we have developed similar high-level models for NVM, including PCRAMsim [Dong 2009], NVSim [Dong 2012], and NVMain [Poremba 2012]. NVSim is a circuit-level model for NVM performance, energy, and area estimation, which supports various NVM technologies, including STT-RAM, PCRAM, ReRAM, and legacy NAND Flash. NVSim is successfully validated against industrial NVM prototypes, and it is expected to help boost architecture-level NVM-related studies. On the other side, NVMain is a cycle accurate main memory simulator designed to simulate emerging nonvolatile memories at the architectural level. We have released these models as open source tools and provided contiguous support to them. We also proposed PS3-RAM, which is a fast, portable and scalable statistical STT-RAM reliability analysis model [Wen 2012]. Design Space Exploration and Optimization With the support of these models, we explore different device/circuit optimization techniques. For example, in [Niu 2012a] we studied the power reduction technique for the application of ECC scheme in ReRAM designs and proposed to use ECC code to relax the BER (Bit Error Rate) requirement of a single memory to improve the write energy consumption and latency for both 1T1R and cross-point ReRAM designs. In [Xu 2011], we proposed a methodology to design STT-RAM for different optimization goals such as read performance, write performance and write energy by leveraging the trade-off between write current and write time of MTJ. We also studied the tradeoffs in building a reliable crosspoint ReRAM array [Niu 2012b]. We have conducted an in depth analysis of the circuit and system level design implications of multi-layer cross-point Resistive RAM (MLCReRAM) from performance, power and reliability perspectives [Xu 2013]. The objective of this study is to understand the design trade-offs of this technology with respect to the MLC Phase Change Memory (MLCPCM).Our MLC ReRAM design at the circuit and system levels indicates that different resistance allocation schemes, programming strategies, peripheral designs, and material selections profoundly affect the area, latency, power, and reliability of MLC ReRAM. Based on this analysis, we conduct two case studies: first we compare MLC ReRAM design against MLC phase-change memory (PCM) and multi-layer cross-point ReRAM design, and point out why multi-level ReRAM is appealing; second we further explore the design space for MLC ReRAM. Architecture and Application We explored hybrid checkpointing using phase-change memory for future exascale systems [Dong 2011] and showed that the use of nonvolatile memory for local checkpointing significantly increases the number of faults covered by local checkpoints and reduces the probability of a global failure in the middle of a global checkpoint to less than 1%. We also proposed a technique called i2WAP to mitigate the write variations in NVM-based last-level cache for the improvement of the NVM lifetime [Wang 2013]. Our wear leveling technique attempts to work around the limitations of write endurance by arranging data access so that write operations can be distributed evenly across all the storage cells. During our intensive research on fault-tolerant NVM design, we found that ECC cannot effectively tolerate hard errors from limited write endurance and process imperfection. Therefore, we devised a novel Point and Discard (PAD) architecture in in [ 2012] as a hard-error-tolerant architecture for ReRAM-based Last Level Caches. PAD improves the lifetime of ReRAM caches by 1.6X-440X under different process variations without performance overhead in the system's early life. We have investigated the applicability of NVM for persistent memory design [Zhao 2013]. New byte addressable NVM enables fast persistent memory that allows in-memory persistent data objects to be updated with much higher throughput. Despite the significant improvement, the performance of these designs is only 50% of the native system with no persistence support, due to the logging or copy-on-write mechanisms used to update the persistent memory. A challenge in this approach is therefore how to efficiently enable atomic, consistent, and durable updates to ensure data persistence that survives application and/or system failures. We have designed a persistent memory system, called Klin, that can provide performance as close as that of the native system. The Klin design adopts a non-volatile cache and a non-volatile main memory for constructing a multi-versioned durable memory system, enabling atomic updates without logging or copy-on-write. Our evaluation shows that the proposed Kiln mechanism can achieve up to 2X of performance improvement to NVRAM-based persistent memory employing write-ahead logging. In addition, our design has numerous practical advantages: a simple and intuitive abstract interface, microarchitecture-level optimizations, fast recovery from failures, and no redundant writes to slow non-volatile storage media. The work was published in MICRO 2013 and received Best Paper Honorable Mentioned Award.
Explore the source record for details and available documents.
Year after year, network traffic keeps reaching newhighs as unprecedented volumes of data flow between an ever-increasing number of sources and destinations. As part of the Analysis on the Wire project, we aim to develop a network-centric approach for streaming data processing to facilitate scientific data analysis and reduce the overhead in sending big data to a data center. This work discusses our design for programmable network devices to augment network capabilities for streaming data processing, including efforts in progress at Brookhaven National Laboratory and the challenges faced to date.