Search NASA⌕ Search

SEARCH · Search NASA

Results for “Fault Tolerant Algorithms”

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 127 records · Page 7

Practical State Machine Replication with Confidentiality

We study how to enable arbitrary randomized algorithms in Byzantine fault-tolerant (BFT) settings. We formalize a randomized BFT protocol and provide a simple and efficient construction that can be built on any existing BFT protocols while adding practically no overhead. We go one step further to revisit a confidential BFT protocol (Yin et al., SOSP '03). We show that their scheme is potentially susceptible to safety and confidentiality attacks. We then present a new protocol that is secure in the stronger model we formalize, by extending the idea of a randomized BFT protocol. Our protocol uses only efficient symmetric cryptography, while Yin et al.'s uses costly threshold signatures. We implemented and evaluated our protocols on microbenchmarks and real-world use cases. We show that our randomized BFT protocol is as efficient as conventional BFT protocols, and our confidential BFT protocol is two to three orders of magnitude faster than Yin et al.'s, which is less secure than ours.

Duan, Sisi↗

A validation methodology for fault-tolerant clock synchronization

A validation method for the synchronization subsystem of a fault-tolerant computer system is presented. The high reliability requirement of flight crucial systems precludes the use of most traditional validation methods. The method presented utilizes formal design proof to uncover design and coding errors and experimentation to validate the assumptions of the design proof. The experimental method is described and illustrated by validating an experimental implementation of the Software Implemented Fault Tolerance (SIFT) clock synchronization algorithm. The design proof of the algorithm defines the maximum skew between any two nonfaulty clocks in the system in terms of theoretical upper bounds on certain system parameters. The quantile to which each parameter must be estimated is determined by a combinatorial analysis of the system reliability. The parameters are measured by direct and indirect means, and upper bounds are estimated. A nonparametric method based on an asymptotic property of the tail of a distribution is used to estimate the upper bound of a critical system parameter. Although the proof process is very costly, it is extremely valuable when validating the crucial synchronization subsystem.

Johnson, S. C.↗

Machine-checked proofs of the design and implementation of a fault-tolerant circuit

A formally verified implementation of the 'oral messages' algorithm of Pease, Shostak, and Lamport is described. An abstract implementation of the algorithm is verified to achieve interactive consistency in the presence of faults. This abstract characterization is then mapped down to a hardware level implementation which inherits the fault-tolerant characteristics of the abstract version. All steps in the proof were checked with the Boyer-Moore theorem prover. A significant results is the demonstration of a fault-tolerant device that is formally specified and whose implementation is proved correct with respect to this specification. A significant simplifying assumption is that the redundant processors behave synchronously. A mechanically checked proof that the oral messages algorithm is 'optimal' in the sense that no algorithm which achieves agreement via similar message passing can tolerate a larger proportion of faulty processor is also described.

Bevier, William R.↗

Hypothetical Scenario Generator for Fault-Tolerant Diagnosis

The Hypothetical Scenario Generator for Fault-tolerant Diagnostics (HSG) is an algorithm being developed in conjunction with other components of artificial- intelligence systems for automated diagnosis and prognosis of faults in spacecraft, aircraft, and other complex engineering systems. By incorporating prognostic capabilities along with advanced diagnostic capabilities, these developments hold promise to increase the safety and affordability of the affected engineering systems by making it possible to obtain timely and accurate information on the statuses of the systems and predicting impending failures well in advance. The HSG is a specific instance of a hypothetical- scenario generator that implements an innovative approach for performing diagnostic reasoning when data are missing. The special purpose served by the HSG is to (1) look for all possible ways in which the present state of the engineering system can be mapped with respect to a given model and (2) generate a prioritized set of future possible states and the scenarios of which they are parts.

James, Mark↗

Demonstration of qubit operations below a rigorous fault tolerance threshold with gate set tomography

Quantum information processors promise fast algorithms for problems inaccessible to classical computers. But since qubits are noisy and error-prone, they will depend on fault-tolerant quantum error correction (FTQEC) to compute reliably. Quantum error correction can protect against general noise if—and only if—the error in each physical qubit operation is smaller than a certain threshold. The threshold for general errors is quantified by their diamond norm. Until now, qubits have been assessed primarily by randomized benchmarking, which reports a different error rate that is not sensitive to all errors, and cannot be compared directly to diamond norm thresholds. Finally, we use gate set tomography to completely characterize operations on a trapped-Yb + -ion qubit and demonstrate with greater than 95% confidence that they satisfy a rigorous threshold for FTQEC (diamond norm ≤6.7 × 10 -4 ).

97 MATHEMATICS AND COMPUTING↗

Communications protocols for a fault tolerant, integrated local area network for Space Station applications

The evolutionary growth of the Space Station and the diverse activities onboard are expected to require a hierarchy of integrated,local area networks capable of supporting data, voice and video communications. In addition, fault tolerant network operation is necessary to protect communications between critical systems attached to the net and to relieve the valuable human resources onboard Space Station of day-to-day data system repair tasks. An experimental, local area network is being developed which will serve as a testbed for investigating candidate algorithms and technologies for a fault tolerant, integrated network. The establishment of a set of rules or protocols which govern communications on the net is essential to obtain orderly and reliable operation. A hierarchy of protocols for the experimental network is presented and procedures for data and control communications are described.

Meredith, B. D.↗

Towards provably efficient quantum algorithms for large-scale machine-learning models

Large machine learning models are revolutionary technologies of artificial intelligence whose bottlenecks include huge computational expenses, power, and time used both in the pre-training and fine-tuning process. In this work, we show that fault-tolerant quantum computing could possibly provide provably efficient resolutions for generic (stochastic) gradient descent algorithms, scaling as $\mathcal{O}$(T 2 x polylog($n$)), where n is the size of the models and T is the number of iterations in the training, as long as the models are both sufficiently dissipative and sparse, with small learning rates. Based on earlier efficient quantum algorithms for dissipative differential equations, we find and prove that similar algorithms work for (stochastic) gradient descent, the primary algorithm for machine learning. In practice, we benchmark instances of large machine learning models from 7 million to 103 million parameters. We find that, in the context of sparse training, a quantum enhancement is possible at the early stage of learning after model pruning, motivating a sparse parameter download and re-upload scheme. Our work shows solidly that fault-tolerant quantum algorithms could potentially contribute to most state-of-the-art, large-scale machine-learning problems.

97 MATHEMATICS AND COMPUTING↗

Design and Implementation of a Distributed Version of the NASA Engine Performance Program

Distributed NEPP is a new version of the NASA Engine Performance Program that runs in parallel on a collection of Unix workstations connected through a network. The program is fault-tolerant, efficient, and shows significant speed-up in a multi-user, heterogeneous environment. This report describes the issues involved in designing distributed NEPP, the algorithms the program uses, and the performance distributed NEPP achieves. It develops an analytical model to predict and measure the performance of the simple distribution, multiple distribution, and fault-tolerant distribution algorithms that distributed NEPP incorporates. Finally, the appendices explain how to use distributed NEPP and document the organization of the program's source code.

Cours, Jeffrey T.↗

Algorithms and Libraries

This exploratory study initiated our inquiry into algorithms and applications that would benefit by latency tolerant approach to algorithm building, including the construction of new algorithms where appropriate. In a multithreaded execution, when a processor reaches a point where remote memory access is necessary, the request is sent out on the network and a context--switch occurs to a new thread of computation. This effectively masks a long and unpredictable latency due to remote loads, thereby providing tolerance to remote access latency. We began to develop standards to profile various algorithm and application parameters, such as the degree of parallelism, granularity, precision, instruction set mix, interprocessor communication, latency etc. These tools will continue to develop and evolve as the Information Power Grid environment matures. To provide a richer context for this research, the project also focused on issues of fault-tolerance and computation migration of numerical algorithms and software. During the initial phase we tried to increase our understanding of the bottlenecks in single processor performance. Our work began by developing an approach for the automatic generation and optimization of numerical software for processors with deep memory hierarchies and pipelined functional units. Based on the results we achieved in this study we are planning to study other architectures of interest, including development of cost models, and developing code generators appropriate to these architectures.

Dongarra, Jack↗

Scalar quantum field theories as a benchmark for near-term quantum computers

Quantum field theory (QFT) simulations are a potentially important application for noisy intermediate scale quantum (NISQ) computers. The ability of a quantum computer to emulate a QFT therefore constitutes a natural application-centric benchmark. Foundational quantum algorithms to simulate QFT processes rely on fault-tolerant computational resources, but to be useful on NISQ machines, error-resilient algorithms are required. Here we outline and implement a hybrid algorithm to calculate the lowest energy levels of the paradigmatic 1+1–dimensional $\phi$ 4 interacting scalar QFT. We calculate energy splittings and compare results with experimental values obtained on currently available quantum hardware. We show that the accuracy of mass-renormalization calculations represents a useful metric with which near-term hardware may be benchmarked. Finally, we also discuss the prospects of scaling the algorithm to full simulation of interacting QFTs on future hardware.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Tolerating Correlated Failures for Generalized Cartesian Distributions via Bipartite Matching

Faults are expected to play an increasingly important role in how algorithms and applications are designed to run on future extreme-scale systems. A key ingredient of any approach to fault tolerance is effective support for fault tolerant data storage. A typical application execution consists of phases in which certain data structures are modified while others are read-only. Often, read-only data structures constitute a large fraction of total memory consumed. Fault tolerance for read-only data can be ensured through the use of checksums or parities, without resorting to expensive in-memory duplication or checkpointing to secondary storage. In this paper, we present a graph-matching approach to compute and store parity data for read-only matrices that are compatible with fault tolerant linear algebra (FTLA). Typical approaches only support blocked data distributions with each process holding one block with the parity located on additional processes. The matrices are assumed to be blocked by a cartesian grid with each block assigned to a process. We consider a generalized distribution in which each process can be assigned arbitrary blocks. We also account for the fact that multiple processes might be part of the same failure unit, say an SMP node. The flexibility enabled by our novel application of graph matching extends fault tolerance support to data distributions beyond those supported by prior work. We evaluate the matching implementations and cost to compute the parity and recover lost data, demonstrating the low overhead incurred by our approach.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Simulation and power quality analysis of a Loose-Coupled bipolar DC microgrid in an office building

With distributed generation and battery storage technologies thriving in microgrids, the use of direct current (DC) microgrids in the building sector offers multiple advantages in energy efficiency and power quality compared with alternating current (AC) systems. This study developed a new concept of a loose-coupled bipolar DC building power system. In this work, the concept was used to design a real-world office building in Shenzhen, China. A power system model was developed to study the stability and control of the DC power system and to verify DC power quality. The design and modeling of the DC power control system is discussed in detail. The study developed a few common fault scenarios in DC building microgrids that were simulated in the MATLAB-Simulink environment to validate the design of a loose-coupled bipolar DC system. The results indicate that the proposed loose-coupled bipolar DC system schema, when implemented with proper control algorithms, can achieve good fault-tolerant performance with reliable power quality, even during disruptive system events.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Design for dependability: A simulation-based approach

This research addresses issues in simulation-based system level dependability analysis of fault-tolerant computer systems. The issues and difficulties of providing a general simulation-based approach for system level analysis are discussed and a methodology that address and tackle these issues is presented. The proposed methodology is designed to permit the study of a wide variety of architectures under various fault conditions. It permits detailed functional modeling of architectural features such as sparing policies, repair schemes, routing algorithms as well as other fault-tolerant mechanisms, and it allows the execution of actual application software. One key benefit of this approach is that the behavior of a system under faults does not have to be pre-defined as it is normally done. Instead, a system can be simulated in detail and injected with faults to determine its failure modes. The thesis describes how object-oriented design is used to incorporate this methodology into a general purpose design and fault injection package called DEPEND. A software model is presented that uses abstractions of application programs to study the behavior and effect of software on hardware faults in the early design stage when actual code is not available. Finally, an acceleration technique that combines hierarchical simulation, time acceleration algorithms and hybrid simulation to reduce simulation time is introduced.

Goswami, Kumar K.↗

Software reliability through fault-avoidance and fault-tolerance

Twenty independently developed but functionally equivalent software versions were used to investigate and compare empirically some properties of N-version programming, Recovery Block, and Consensus Recovery Block, using the majority and consensus voting algorithms. This was also compared with another hybrid fault-tolerant scheme called Acceptance Voting, using dynamic versions of consensus and majority voting. Consensus voting provides adaptation of the voting strategy to varying component reliability, failure correlation, and output space characteristics. Since failure correlation among versions effectively reduces the cardinality of the space in which the voter make decisions, consensus voting is usually preferable to simple majority voting in any fault-tolerant system. When versions have considerably different reliabilities, the version with the best reliability will perform better than any of the fault-tolerant techniques.

Vouk, Mladen A.↗

Multi-version software reliability through fault-avoidance and fault-tolerance

A number of experimental and theoretical issues associated with the practical use of multi-version software to provide run-time tolerance to software faults were investigated. A specialized tool was developed and evaluated for measuring testing coverage for a variety of metrics. The tool was used to collect information on the relationships between software faults and coverage provided by the testing process as measured by different metrics (including data flow metrics). Considerable correlation was found between coverage provided by some higher metrics and the elimination of faults in the code. Back-to-back testing was continued as an efficient mechanism for removal of un-correlated faults, and common-cause faults of variable span. Software reliability estimation methods was also continued based on non-random sampling, and the relationship between software reliability and code coverage provided through testing. New fault tolerance models were formulated. Simulation studies of the Acceptance Voting and Multi-stage Voting algorithms were finished and it was found that these two schemes for software fault tolerance are superior in many respects to some commonly used schemes. Particularly encouraging are the safety properties of the Acceptance testing scheme.

Vouk, Mladen A.↗

Control optimization, stabilization and computer algorithms for aircraft applications

The analysis and design of complex multivariable reliable control systems are considered. High performance and fault tolerant aircraft systems are the objectives. A preliminary feasibility study of the design of a lateral control system for a VTOL aircraft that is to land on a DD963 class destroyer under high sea state conditions is provided. Progress in the following areas is summarized: (1) VTOL control system design studies; (2) robust multivariable control system synthesis; (3) adaptive control systems; (4) failure detection algorithms; and (5) fault tolerant optimal control theory.

Athans, M.↗

Petascale Computing Enabling Technologies Project Final Report

The Petascale Computing Enabling Technologies (PCET) project addressed challenges arising from current trends in computer architecture that will lead to large-scale systems with many more nodes, each of which uses multicore chips. These factors will soon lead to systems that have over one million processors. Also, the use of multicore chips will lead to less memory and less memory bandwidth per core. We need fundamentally new algorithmic approaches to cope with these memory constraints and the huge number of processors. Further, correct, efficient code development is difficult even with the number of processors in current systems; more processors will only make it harder. The goal of PCET was to overcome these challenges by developing the computer science and mathematical underpinnings needed to realize the full potential of our future large-scale systems. Our research results will significantly increase the scientific output obtained from LLNL large-scale computing resources by improving application scientist productivity and system utilization. Our successes include scalable mathematical algorithms that adapt to these emerging architecture trends and through code correctness and performance methodologies that automate critical aspects of application development as well as the foundations for application-level fault tolerance techniques. PCET's scope encompassed several research thrusts in computer science and mathematics: code correctness and performance methodologies, scalable mathematics algorithms appropriate for multicore systems, and application-level fault tolerance techniques. Due to funding limitations, we focused primarily on the first three thrusts although our work also lays the foundation for the needed advances in fault tolerance. In the area of scalable mathematics algorithms, our preliminary work established that OpenMP performance of the AMG linear solver benchmark and important individual kernels on Atlas did not match the predictions of our simple initial model. Our investigations demonstrated that a poor default memory allocation mechanism degraded performance. We developed a prototype NUMA library to provide generic mechanisms to overcome these issues, resulting in significantly improved OpenMP performance. After additional testing, we will make this library available to all users, providing a simple means to improve threading on LLNL's production Linux platforms. We also made progress on developing new scalable algorithms that target multicore nodes. We designed and implemented a new AMG interpolation operator with improved convergence properties for very low complexity coarsening schemes. This implementation will also soon be available to LLNL's application teams as part of the hypre library. We presented results for both topics in an invited plenary talk entitled 'Efficient Sparse Linear Solvers for Multi-Core Architectures' at the 2009 HPCMP Institutes Annual Meeting/CREATE Annual All-Hands Meeting. The interpolation work was summarized in a talk entitled 'Improving Interpolation for Aggressive Coarsening' at the 14th Copper Mountain Conference on Multigrid Methods and in a research paper that will appear in Numerical Linear Algebra with Applications. In the area of code correctness, we significantly extended our behavior equivalence class identification mechanism. Specifically, we not only demonstrated it works well at very large scales but we also added the ability to classify MPI tasks not only by function call traces, but also by specific call sites (source code line numbers) being executed by tasks. More importantly, we developed a new technique to determine relative logical execution progress of tasks in the equivalence classes by combining static analysis with our original dynamic approach. We applied this technique to a correctness issue that arose at 4096 tasks during the development of the new AMG interpolation operator discussed above. This scale isat the limit of effectiveness of production tools, but our technique quickly located the erroneous source code, demonstrating the power of understanding relationships between behavioral equivalence classes. This work is the subject of a paper recently accepted to SC09, as well as a presentation entitled 'Providing Order to Extreme Scale Debugging Chaos' given at the ParaDyn Week annual conference in College Park, MD. In addition to this theoretical extension, we have made significant progress in developing a front end for this tool set, and the front-end is now available on several of LLNL's largescale computing resources. In addition, we explored mechanisms to identify exact locations of erroneous MPI usage in application source code. In this work, we developed a new model that led to a highly efficient algorithm for detecting deadlock during dynamic software testing. This work was the subject of a well-received paper at ICS 2009 [4].

99 GENERAL AND MISCELLANEOUS↗