The Hydra Framework is a new, principled approach to modeling and detecting security-critical bugs. Fusing a variant of classical n-version (redundant) programming with automated bug bounty payouts, Hydra provides eco...
详细信息
The Hydra Framework is a new, principled approach to modeling and detecting security-critical bugs. Fusing a variant of classical n-version (redundant) programming with automated bug bounty payouts, Hydra provides economically rigorous and cost-effective bounty protections for smart contracts.
Quality control of machine learning systems is a fundamental challenge in industries to provide intelligent services or products using machine learning. While recent advances in machine learning algorithms substantial...
详细信息
ISBN:
(纸本)9781728130309
Quality control of machine learning systems is a fundamental challenge in industries to provide intelligent services or products using machine learning. While recent advances in machine learning algorithms substantially improve the performance of intelligent tasks such as object recognition, their outputs are essentially stochastic and very sensitive to input data. Such an output uncertainty is a big obstacle to ensure the quality of safety critical applications like autonomous vehicle and hence architectural design to mitigate the impact of error output becomes a great importance. In this paper, we propose n-version machine learning architecture that aims to improve system reliability against probabilistic outputs of individual machine learning modules. The key idea of this architecture is exploiting two kinds of diversities;input diversity and model diversity. Our study first formally defines these diversity metrics and analytically shows the improved reliability by n-version machine learning architecture. Since we treat a machine learning module as a black-box, the proposed architecture and the reliability property are generally applicable to any machine learning algorithms and applications.
Since it was first proposed in 1978, n-version programming (nVP) has been a mainstay technique for obtaining fault tolerance in high-reliability software. Although subsequent research onnVP has indicated that coincid...
详细信息
Since it was first proposed in 1978, n-version programming (nVP) has been a mainstay technique for obtaining fault tolerance in high-reliability software. Although subsequent research onnVP has indicated that coincident failures across different versions can be more common than initially expected, nVP remains the “gold standard” for application domains in which high reliability is of critical importance, such as fly-by-wire avionics control software.
The main issue in controlling safety-critical systems such as nuclear power reactors or railway signalization systems where risk ratio is at the highest level because small errors might result in hazardous accidents (...
详细信息
The main issue in controlling safety-critical systems such as nuclear power reactors or railway signalization systems where risk ratio is at the highest level because small errors might result in hazardous accidents (e.g. death or injury of several people) is to satisfy safety-related standards and provide high-safety with high-reliability. In order to improve reliability and safety of such systems to the required level by decreasing common-cause failures at the same time, diverse programming (or n-version programming) technique where n-different independently designed modules run in parallel on the same input output specifications can be used. In this study, a new bitwise voting strategy based on safe-states of variables is proposed and possible critical races between modules those run in a parallel manner are determined. Solutions for critical races are also explained.
The paper describes and evaluates Fireplug, a flexible architecture to build robust geo-replicated graph databases. Fireplug can be configured to tolerate from crash to Byzantine faults, both within and across differe...
详细信息
ISBN:
(纸本)9781538622902
The paper describes and evaluates Fireplug, a flexible architecture to build robust geo-replicated graph databases. Fireplug can be configured to tolerate from crash to Byzantine faults, both within and across different datacenters. Furthermore, Fireplug is robust to bugs in existing graph database implementations, as it allows to combine multiple graph databases instances in a cohesive manner. Thus, Fireplug can support many different deployments, according to the performance/robustness tradeoffs imposed by the target application. Our evaluation shows that Fireplug can implement Byzantine fault tolerance in geo-replicated scenarios and still outperform the built-in replication mechanism of neo4j, which only supports crash faults.
作者：
Yang, XinLi, HuiWang, HanPeking Univ
Huawei & PKU Jointly Engn Lab Future Network Base Shenzhen Engn Lab Converged Networking TechnolPK Future Network PKU Lab Natl Major Res Infrastruct Shenzhen 518055 Peoples R China
Moving target defense (MTD) techniques are effective solutions to improve the network security by continuously reconfiguring the system setting. On the other hand, continuously transforming also increase the cost of d...
详细信息
ISBN:
(纸本)9781538669501
Moving target defense (MTD) techniques are effective solutions to improve the network security by continuously reconfiguring the system setting. On the other hand, continuously transforming also increase the cost of defenders, so it is important to analyze the effectiveness of MTDs compared with their cost. Current researches lack of analyzing the effectiveness by mathematical theory compared with analyzing by experiment. Motivated by the above, we propose a novel three-dimension model named nPM jointly use n-version programming, Poisson process, Markov chain and martingale theory to analyze the effectiveness of the proposed MTD model. Our analysis points out the difficulty for a successful adversary to defeat the MTD system, which is related to the system configuration, such as the number of executors and the judgment criterion in every node, the transforming period and rang of system MTD transformation. Finally, we give advices on the design of the system in the daily defense and the attacked defense, with the goal of guaranteeing security with minimal cost.
Complex systems-of-systems (SoS) requiring fault-tolerance and high reliability often require redundant systems. The concept of redundancy that includes components with differing failure modes is well understood in th...
详细信息
ISBN:
(纸本)9781424494927
Complex systems-of-systems (SoS) requiring fault-tolerance and high reliability often require redundant systems. The concept of redundancy that includes components with differing failure modes is well understood in the realm of hardware design. n-version programming, although shown to produce significant gains in software reliability over single-version fault tolerant techniques, is not widely accepted or applied. This is due, in part, to n-version programming's lengthy development time and its inherent problems with version independence. Model Driven Software Development (MDSD) is a process that promises gains in software productivity and quality. While progress in MDSD has witnessed the expansion of the supporting Unified Modeling Language profile for modeling fault tolerant characteristics, and the development of specific design patterns for the production of fault tolerant software, MDSD's support in the generation and testing of fault tolerant applications is not evident or explicitly defined. This paper discusses the development of a fault tolerant MDSD framework, which enables users to design, implement and test fault tolerant applications using the n-version modeling technique. The framework closes the gap between existing modeling patterns and the practical application of fault tolerant MDSD, and supports follow-on research to address specific questions relating to the benefits of MDSD within the fault tolerance software domain.
This paper takes an axiomatic and calculational view of diversity (or "n-version programming"), where multiple implementations of the same specification are executed in parallel to increase dependability. Th...
详细信息
This paper takes an axiomatic and calculational view of diversity (or "n-version programming"), where multiple implementations of the same specification are executed in parallel to increase dependability. The central notion is "adjudication": once we have multiple, potential different, outcomes, how do we come to a single result? Adjudication operators are explicitly defined and some general properties for these explored. (C) 2015 Elsevier Inc. All rights reserved.
Presently, service-based software systems ( SBSSs) have been heavily deployed to fulfil the functionalities of cloud computing and are widely used in many other application fields. Additionally, maintaining functional...
详细信息
Presently, service-based software systems ( SBSSs) have been heavily deployed to fulfil the functionalities of cloud computing and are widely used in many other application fields. Additionally, maintaining functionality and quality of service levels becomes increasingly important for SBSSs;this is because system operational failures may cause great financial loss to an organization. Fault tolerance ( FT) is usually used to provide continuous and reliable system service delivery when failures occur. However, the reliability and performance of FT should be carefully analysed because of the overhead of invoking redundant services. It is also noted that the single point of failure on the FT adjudicators as well as the failure correlation also hamper the benefits of FT in SBSSs. To address these problems, this paper proposes two approaches, the stochastic modelling approach and the simulation approach, for analysing the reliability and performance of generalized FT designs. The first approach is suitable for quick analysis at an early design stage, while the second approach is built on top of the ns-3 simulator and could be well adapted to incorporate varied uncertainty models in the SBSS environments Extensive experiments and analyses uncover some characteristics that could be useful for SBSS engineers. Copyright (C) 2015 John Wiley & Sons, Ltd.
The widespread reliance on software for mission and life critical applications makes the reliability of these systems essential. Techniques such as fault tolerance have been proposed to achieve the highest levels of s...
详细信息
The widespread reliance on software for mission and life critical applications makes the reliability of these systems essential. Techniques such as fault tolerance have been proposed to achieve the highest levels of software reliability. However, the fault tolerance paradigm suffers from the risk of correlated failures, where a majority of the software versions fail on the same input leading to system failure. This paper derives a trivariate Bernoulli distribution to quantify the negative impact of correlated failures on the reliability of fault tolerant software composed of highly reliable versions. An experiment based on early empirical research demonstrates the capacity of the distribution to conduct reliability assessment for many combinations of the version reliabilities and correlations. The results indicate that correlated failures detract from system reliability, but that this reliability is often higher than a system composed of the single most reliable version.
暂无评论