SEARCH · Search NASA
Results for “system scaling”
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.
Rolex: Resilience-oriented language extensions for extreme-scale systems
Future exascale high-performance computing (HPC) systems will be constructed from VLSI devices that will be less reliable than those used today, and faults will become the norm, not the exception. This will pose significant problems for system designers and programmers, who for half-a-century have enjoyed an execution model that assumed correct behavior by the underlying computing system. The mean time to failure (MTTF) of the system scales inversely to the number of components in the system and therefore faults and resultant system level failures will increase, as systems scale in terms of the number of processor cores and memory modules used. However every error detected need not cause catastrophic failure. Many HPC applications are inherently fault resilient. Yet it is the application programmers who have this knowledge but lack mechanisms to convey it to the system. In this paper, we present new Resilience Oriented Language Extensions (Rolex) which facilitate the incorporation of fault resilience as an intrinsic property of the application code. We describe the syntax and semantics of the language extensions as well as the implementation of the supporting compiler infrastructure and runtime system. Furthermore, our experiments show that an approach that leverages the programmer's insight to reason about the context and significance of faults to the application outcome significantly improves the probability that an application runs to a successful conclusion.
Thermal runaway propagation models: from module scale to system scale.
Abstract not provided.
Sentiment Analysis based Error Detection for Large-Scale Systems
Today's large-scale systems such as High Performance Computing (HPC) Systems are designed/utilized towards exascale computing, inevitably decreasing its reliability due to the increasing design complexity. HPC systems conduct extensive logging of their execution behaviour. In this paper, we leverage the inherent meaning behind the log messages and propose a novel sentiment analysis-based approach for the error detection in large-scale systems, by automatically mining the sentiments in the log messages. Our contributions are four-fold. (1) We develop a machine learning (ML) based approach to automatically build a sentiment lexicon, based on the system log message templates. (2) Using the sentiment lexicon, we develop an algorithm to detect system errors. (3) We develop an algorithm to identify the nodes and components with erroneous behaviors, based on sentiment polarity scores. (4) We evaluate our solution vs. other state-of-the-art machine/deep learning algorithms based on three representative supercomputers' system logs. Experiments show that our error detection algorithm can identify error messages with an average MCC score and f -score of 91% and 96% respectively, while state of the art ML/deep learning model (LSTM) obtains only 67% and 84%. To the best of our knowledge, this is the first work leveraging the sentiments embedded in log entries of large-scale systems for system health analysis.
On large-scale system stability.
A large-scale system is considered as a system constituted of subsystems which may be connected or disconnected from each other during operation. A new notion of connective stability is introduced by which a large-scale system is regarded as stable if it remains stable (in the sense of Liapunov) under structural perturbations produced by the on-off participation of the subsystems. Algebraic conditions are developed that guarantee exponential connective stability in large-scale systems which may be composed of linear and nonlinear subsystems coupled by linear or nonlinear connections.
Survey on large scale system control methods
The problem inherent to large scale systems such as power network, communication network and economic or ecological systems were studied. The increase in size and flexibility of future spacecraft has put those dynamical systems into the category of large scale systems, and tools specific to the class of large systems are being sought to design control systems that can guarantee more stability and better performance. Among several survey papers, reference was found to a thorough investigation on decentralized control methods. Especially helpful was the classification made of the different existing approaches to deal with large scale systems. A very similar classification is used, even though the papers surveyed are somehow different from the ones reviewed in other papers. Special attention is brought to the applicability of the existing methods to controlling large mechanical systems like large space structures. Some recent developments are added to this survey.
Stability of large-scale systems under structural perturbations.
A large-scale system is considered as a system constituted of subsystems which may be connected or disconnected from each other during operation. A new concept of connective stability is introduced by which a large-scale system is regarded as stable if it remains stable (in the sense of Lyapunov) under structural perturbations produced by the on-off participation of the subsystems. Algebraic conditions are developed that guarantee exponential connective stability of large-scale systems which may be composed of linear and nonlinear time-varying subsystems coupled by linear or nonlinear connections.
Dynamical System Scaling of a Thermocline Thermal Storage System in the Thermal Energy Distribution System (TEDS) Facility
The purpose of this study was to develop a process to convert input signals from one facility into another by reflecting geometric and environmental settings. The Dynamic Energy Transport and Integration Laboratory (DETAIL) is one facility in development that aims to emulate the daily interactions among power production industry systems and be capable of receiving real-time data from those systems as inputs. To convert signals and ensure that the temporal sequences and magnitudes reflect the laboratory settings, the ability to scale and project data is essential. To demonstrate this ability, Dynamical System Scaling (a methodology that enables systems to scale and project or extrapolate datasets to desired environments while conserving the observed transient behavior based on first principles) was applied to DETAIL’s thermocline thermal storage system in the Thermal Energy Distribution System. The thermocline system was successfully scaled and a test case was conducted to generate a doubly accelerated energy charge and discharge in reference to past experimental data from the facility. The accelerated data were determined as able to conserve the amount of energy stored and the associated test boundary conditions were charge line maximum temperature, charge line velocity, and thermocline maximum temperature at 354 °C, 0.458 m/s, and 418 °C, respectively. The research results represented a case that required signals to be accelerated without altering the stored energy.
A bibliographical survey of large-scale systems
Bibliography on large scale systems such as large societal, communication, vehicular traffic control, and management systems
Application of Dynamical System Scaling to Bubble Dynamics
Dynamical System Scaling (DSS) provides a useful method for analyzing, categorizing and scaling time-dependent processes. A key feature of DSS is the temporal displacement rate, D, which relates the natural process time to the reference clock time. Because of its property of being invariant under a two-parameter affine transformation, it provides an underlying basis for process scaling. Application of DSS to bubble dynamics serves to elucidate the physical significance of the temporal displacement rate and its role in scaling a variety of bubble dynamics processes. This paper shows that the temporal displacement rate consists of the sum of dimensionless groups that govern the bubble dynamics. Preserving the temporal displacement rate for a prototype and a scaled model results in similitude of the time-dependent normalized bubble size, growth rates and interfacial accelerations for different fluid conditions. For vapor bubble growth in a superheated liquid, it is shown that inertial controlled bubble growth occurs when D=0 and thermally controlled bubble growth occurs when D=1 .
SharP Hash: A High-Performing Distributed Hash for Extreme-Scale Systems
A high-performing distributed hash is critical for achieving performance in many applications and system software using extreme-scale systems. It is also a central part of many Big-Data frameworks including Memcached, file systems, and job schedulers. However, there is a lack of high-performing distributed hash implementations. In this work, we propose, design, and implement, SharP Hash, a high-performing, RDMA-based distributed hash for extreme-scale systems. SharP Hash's high performance is obtained through the use of high-performing networks and one-sided semantics. We perform an evaluation of SharP Hash and demonstrate its performance characteristics with a synthetic micro-benchmark and implementation of a Key Value (KV) store, Memcached.
Emulation Framework for Distributed Large-Scale Systems Integration
Recent trends in systems engineering include integration of very large-scale systems, which entails significant challenges when they are geographically dispersed. In these scenarios, intelligent integration of distributed large-scale systems requires significant coordination among hardware elements as well as all software components. The approach of integrated systems (both computing platform and experimental equipment) for end-to-end orchestration is called federation. Virtual frameworks can aid in the testing, assessment, and implementation of a functional system of interconnected resources. We present an emulation framework that replicates the software environments of multi-site federations of computing systems and instruments. Our emulation framework allows systems engineers to reduce developmentcost and avoid disruptions to production infrastructure. Our framework was effectively used to develop and test software modules for various tasks including container orchestration and instrument access. For performance assessment, however, the emulated framework is severely limited in providing accurate network and IO measurements at 10 Gbps and higher data rates. The data transfer performance profiles estimated using these emulated measurements are usually inaccurate for high bandwidth and high latency connections, since emulation does not accurately reflect the critical network transport dynamics.We utilize measurements from a physical testbed with hardware network emulators to obtain data transfer profiles that closely match the expected profiles for the emulated federations. We show the effectiveness of our approach by an illustrative example of integrated (federated) multi-site ultra large-scale systems that are connected via high speed wide area networks.
Machine Learning-Based Model Predictive Control of Two-Time-Scale Systems
In this study, we present a general form of nonlinear two-time-scale systems, where singular perturbation analysis is used to separate the dynamics of the slow and fast subsystems. Machine learning techniques are utilized to approximate the dynamics of both subsystems. Specifically, a recurrent neural network (RNN) and a feedforward neural network (FNN) are used to predict the slow and fast state vectors, respectively. Moreover, we investigate the generalization error bounds for these machine learning models approximating the dynamics of two-time-scale systems. Next, under the assumption that the fast states are asymptotically stable, our focus shifts toward designing a Lyapunov-based model predictive control (LMPC) scheme that exclusively employs the RNN to predict the dynamics of the slow states. Additionally, we derive sufficient conditions to guarantee the closed-loop stability of the system under the sample-and-hold implementation of the controller. A nonlinear chemical process example is used to demonstrate the theory. In particular, two RNN models are constructed: one to model the full two-time-scale system and the other to predict solely the slow state vector. Both models are integrated within the LMPC scheme, and we compare their closed-loop performance while assessing the computational time required to execute the LMPC optimization problem.
Supporting Co-Design of Extreme-Scale Systems with In Situ Visual Analysis of Event-Driven Simulations (Final Report)
Extreme-scale systems for compute- and data-centric science are pushing the boundaries of system architectures in order to achieve ambitious goals in terms of computing capability, resilience, and power efficiency. Future systems necessitate emerging designs that incorporate new technologies, system organizations, and computer science techniques, and these changes are not yet well understood. Discrete-event-driven simulation of computing system architectures and subsystems has emerged as a productive and cost-effective means to evaluating potential designs, along with capabilities for executing these simulations of extreme-scale systems. However, techniques for analyzing the behavior of these simulations have not kept pace with our ability to perform the simulations, preventing us from extracting the most value from this promising approach. The objective of this project is to support the codesign of extreme-scale system architectures for compute and data-centric science through research and development of novel methods for analysis and visualization of large-scale event-driven simulations.
Ceramic Membrane Solar Desal: Pilot System Design and Full-Scale System Techno-Economic Analysis (CRADA Final Report)
This work is conducted in support of the American Made Challenges Solar Prize. The current scope addresses the Design portion of the overall prize competition. This team, led by the University of Connecticut, will design a solar powered, pilot scale, ceramic-based membrane distillation system that can operate on high salinity waters and propose that design for construction in the Test phase of the Prize competition.
A Co-Simulation Framework for Steady-State Analyses of Multiple Droop-based MTdc Grids in Continental-Scale Systems
The growing scale and complexity of planning continental hybrid ac and multi-terminal dc (MTdc) systems require scalable steady-state modeling and analysis approaches not currently available in commercial tools. This paper presents a comprehensive multi-fidelity model-conversion framework that enables the efficient transition of MTdc grid models from production cost modeling (PCM) and approximated ac power flow to detailed ac–MTdc power flow for large-scale planning studies. The core of this framework is a scalable co-simulation approach that, for the first time, enables power flow analysis in continental-scale ac–MTdc systems. It seamlessly couples commercial ac solvers with a detailed MTdc grid model that incorporates droop-based control and current-limiting strategies of multiple meshed MTdc grids. Leveraging this capability, an evaluation framework to systematically assess and compare different MTdc power redispatch strategies under ac and dc contingencies is introduced. The proposed framework and algorithm are evaluated using a combined Western and Eastern Interconnection system with 11 MTdc grids of various sizes, showing a coherent transition from PCM to detailed ac–MTdc power flow and improved system performance in voltage regulation and line overload mitigation following typical contingencies.
Using Rollback Avoidance to Mitigate Failures in Next-Generation Extreme-Scale Systems
High-performance computing (HPC) systems enable scientists to numerically model complex phenomena in many important physical systems. The next major milestone in the development of HPC systems is the construction of the rst supercomputer capable executing more than an exa op, 10 18 oating point operations per second. On systems of this scale, failures will occur much more frequently than on current systems. As a result, resilience is a key obstacle to building next-generation extremescale systems. Coordinated checkpointing is currently the most widely-used mechanism for handling failures on HPC systems. Although coordinated checkpointing remains e ective on current systems, increasing the scale of today's systems to build next-generation systems will increase the cost of fault tolerance as more and more time is taken away from the application to protect against or recover from failure. Rollback avoidance techniques seek to mitigate the cost of checkpoint/restart by allowing an application to continue its execution rather than rolling back to an earlier checkpoint when failures occur. These techniqes include failure prediction and preventive migration, replicated computation, fault-tolerant algorithms, and softwarebased memory fault correction. In this thesis, we examine how rollback avoidance techniques can be used to address failures on extreme-scale systems. Using a combination of analytic modeling and simulation, we evaluate the potential impact of rollback avoidance on these systems. We then present a novel rollback avoidance technique that exploits similarities in application memory. Finally, we examine the feasibility of using this technique to protect against memory faults in kernel memory.
Systems analysis methodology for large scale systems
The objective is to develop a systematic approach for defining and evaluating alternative strategies for developing data processing and management systems. Various techniques and systems engineering tools are integrated to develop and present to the analysts and users an early definition of services desired along with probable and compatible operations concepts. Information is presented in the form of outlines, graphs and charts.