Recovery Block Research Articles

Optimal structure of fault-tolerant software systems

This paper considers software systems consisting of fault-tolerant components. These components are built from functionally equivalent but independently developed versions characterized by different reliability and execution time. Because of hardware resource constraints, the number of versions that can run simultaneously is limited. The expected system execution time and its reliability (defined as probability of obtaining the correct output within a specified time) strictly depend on parameters of software versions and sequence of their execution. The system structure optimization problem is formulated in which one has to choose software versions for each component and find the sequence of their execution in order to achieve the greatest system reliability subject to cost constraints. The versions are to be chosen from a list of available products. Each version is characterized by its reliability, execution time and cost. The suggested optimization procedure is based on an algorithm for determining system execution time distribution that uses the moment generating function approach and on the genetic algorithm. Both N-version programming and the recovery block scheme are considered within a universal model. Illustrated example is presented.

Memory of Learning Facilitates Saccadic Adaptation in the Monkey

A motor learning mechanism called saccadic adaptation ensures accuracy of saccades throughout life despite growth, aging, and some pathologies of the oculomotor plant or nervous system. The present study investigates effects of preceding adaptation on the speed of subsequent adaptation during single experiments. Adaptive changes in gain (movement size divided by target eccentricity) were induced by intrasaccadic step (ISS) of the target. After the gain was altered (control block), we reversed the direction of ISS to bring the gain back to approximately 1.0 (recovery). We then reversed ISS direction again to induce another adaptation (test block). Analyses revealed that the gain changed at a higher rate in the early part of test adaptation than in the corresponding part of control. After approximately 100-300 saccades in the test block, adaptation slowed down. The gain value at which adaptation slowed was correlated with the gain achieved in the control. We further examined effects of a 30 min intervention inserted between recovery and test blocks. When zero-visual-error trials ( approximately 700 saccades) were repeated during this period, the rate of test adaptation was similar to that of control. In contrast, when the animal was deprived of visual inputs during this period, test adaptation was still influenced by preceding learning. We conclude that a memory of previous learning remains during recovery to facilitate subsequent adaptation and that such a memory does not disappear merely with time but is erased actively by repeated zero-error movements. Our results, which cannot be explained by a single mechanism, suggest that the saccadic system is equipped with more than one plasticity process.

Merging a DSP-Oriented Signal Integrity Technique and SW-Based Fault Handling Mechanisms to Ensure Reliable DSP Systems

Hereafter, we present an approach aiming to improve the reliability of digital signal processing (DSP) systems operating in real noisy (electromagnetic interference—EMI) environments. The approach is based on the coupling of two techniques: the “DSP-oriented signal integrity improvement” technique deals with increasing the signal-to-noise ratio (SNR) and is essentially a modification of the classic Recovery Blocks Scheme. The second technique, named “SW-based fault handling” aims to detect in real-time data- and control-flow faults throughout modifications of the processor code. When compared to conventional approaches using Fast Fourier Transform (FFT) and Hamming Code, the primary benefit of such an approach is to improve system reliability by means of a considerably low complexity, reasonably low performance degradation and, when implemented in hardware, with reduced area overhead. Aiming to illustrate the proposed approach, we implemented a HW/SW prototype to operate as a speech recognition system (SRS). This prototype was tested under a home-tailored EMI environment according to the IEC 61000-4-29 International Standard Normative. The obtained results indicate that the proposed approach can effectively improve the reliability of DSP systems operating in real noise (EMI) environments.

Reliability and performance analysis for fault-tolerant programs consisting of versions with different characteristics

Abstract This paper presents a simple straightforward algorithm for evaluating reliability and expected execution time for software systems consisting of fault-tolerant components. The components are built from functionally equivalent but independently developed versions characterized by different reliability and performance. Both N-version programming (with parallel and sequential execution of the versions) and the recovery block scheme are considered within a universal model.

Drug binding to HERG channels: evidence for a 'non-aromatic' binding site for fluvoxamine.

British Journal of Pharmacology (2003) 139, 883–884. doi:10.1038/sj.bjp.0705336

The use of neuromuscular monitoring in Germany

As there are no reliable epidemiological data for the use of muscle relaxants in Germany,we conducted a mailing study. The aim of the study was to compare the use of muscle relaxants between German anaesthesia departments. In the present part of the presentation we focused on neuromuscular monitoring (NMM) and management of residual paralysis. A total number of 2,996 questionnaires were sent to all registered anaesthesiological facilities in Germany and the return was 68.6% (2054 questionnaires). In 574 of the returned questionnaires (28%) the regular use of NMM was confirmed. Intraoperative monitoring of neuromuscular block and assessment of neuromuscular recovery were the most frequent applications of NMM, i.e. 25% and 18% of returned questionnaires, respectively. Clinical signs, however, are still the most popular way to estimate the degree of neuromuscular blockade. Moreover, routine reversal at the end of surgery with a neostigmin/atropine mixture was not practiced in 75% of the anaesthesia departments. This survey revealed that NMM is still very rarely used in daily clinical practice. Especially the seldom use of NMM to assess residual paralysis has to be improved.

Numerical model of the dynamic absorption variation in QW-EAM for ultrafast all-optical signal processing

Noise reduction techniques, such as all-optical 3R regeneration, can be realised using ultrafast saturable absorbers, both for the clock recovery block and for the reshaping block. Electroabsorption modulators (EAM) can be used as saturable absorbers when they are traversed by a strong pump optical pulse, which can reduce the absorption experienced by a second probe signal that propagates along the device together with the pump: this feature is called cross-absorption modulation (XAM). The authors present a novel numerical model to evaluate the dynamic behaviour of a quantum well EAM absorption, that can be applied in all-optical signal processing schemes for ultrafast transmission systems. The effects of both electric field screening and exciton saturation are taken into account. Numerical results agree with the experimental device behaviour reported in the literature.

A New On-Line Robust Approach to Design Noise-Immune Speech Recognition Systems

Hereafter, we present a new approach dealing to cope with the harmful effects of noise on speech recognition systems (SRS). This approach is oriented to hardware redundancy and it is essentially a modification of the classic Recovery Blocks scheme. When compared to conventional approaches using Fast Fourier Transform (FFT) and Hamming Code, the primary benefit of such a technique is to improve system performance when operating in real (i.e., noisy) environments. The second advantage is related to the considerably low complexity and reduced area overhead required for implementation. We implemented three full versions of the proposed algorithm: one running of a PC microcomputer, and a second one slightly modified to run on a TMS-320C67 Texas DSP microprocessor module. Both of them were described in the C language. Finally, a last implementation was prototyped on a HW-SW development environment based on the same Texas microprocessor and on the FLEX10K20 FPGA Altera Component.

Design and Verification of Distributed Recovery Blocks with CSP

A case study on the application of Communicating Sequential Processes (CSP) to the design and verification of fault-tolerant real-time systems is presented. The distributed recovery block (DRB) scheme is a design technique for the uniform treatment of hardware and software faults in real-time systems. Through a simple fault-tolerant real-time system design using the DRB scheme, the case study illustrates a paradigm for specifying fault-tolerant software and demonstrates how the different behavioural aspects of a fault-tolerant real-time system design can be separately and systematically specified, formulated, and verified using an integrated set of formal techniques based on CSP.

A new analytical model of shared backup path provisioning in GMPLS networks

As GMPLS and its supporting set of protocols develop into aviable control plane for optical networks, an important function that they will need to support will be the protection and restoration function that has been a major feature of legacy optical networks. A network with a robust set of protection and restoration mechanisms will be able to support data traffic while allowing faster recovery from failures than can be obtained using layer 3 rerouting. Several models have been proposed for protection with GMPLS using shared backup paths. This previous work has not investigated the effect on recovery time critical to the service or the number of backup paths that are required to meet a desired level of performance. Using both restoration time and recovery blocking probability, we have developed a new analytic model for GMPLS-based recovery in M : N protection groups. Furthermore, we show that smaller backup paths can be reserved by capturing the effect of multiple failures in the case of M : N shared protection with revertive mode in an optical network with a GMPLS control plane.

Reliability analysis of N‐version programming with deadline mechanism

In this paper reliability models for fault tolerant N‐version programming and consensus recovery block (combination of N‐version programming and recovery block) with deadline mechanism are analysed. Explicit expressions for reliability of N‐version programming and consensus recovery block with exponential execution times are derived. The paper also presents optimisation models for real time N‐version programming. The objective of the optimisation problem is to maximise reliability of the software satisfying a budget constraint. The paper also includes efficient branch and bound procedure that can be used to solve the optimisation problem.

State restoration in Ada 95: a portable approach to supporting software fault tolerance

Studies indicate that techniques for tolerating hardware faults are so effective that software design errors are the leading cause of all faults encountered. To handle these unanticipated software faults, two main approaches have been proposed: N-version programming and recovery blocks. Both are based on the concept of design diversity: the assumption that different designs will exhibit different faults (if any) for the same inputs and will, therefore, provide alternatives for each other. Both approaches have advantages, but this paper focuses upon recovery blocks; specifically, the requirement to save and restore application state. Judicious saving of state has been described as “checkpointing” for over a decade. Using the object-oriented features of the revised Ada language (Ada 95) – a language widely used in this domain – we present three portable implementations of a checkpointing facility and discuss the trade-offs offered by each. Results of the implementation of these mechanisms are used to highlight both the strengths and weaknesses of some of the object-oriented features of Ada. We then show a reusable implementation of recovery blocks illustrating the checkpointing schemes. A performance analysis is made and measurements are presented in support of the analysis.

SSL/TLS Protocol Enablement for Key Recovery

This document describes an approach for modifying the SSL/TLS (TLS) protocol to allow transmission of Key Recovery Block (KRBs) before communication takes place between clients and servers using strong encryption techniques. From a TLS perspective “strong encryption” implies use of a session key with the secret portion being larger than 40 bits.

A Common Key Recovery Block Format: Promoting Interoperability Between Dissimilar Key Recovery Mechanisms

When cryptographic products that employ different key recovery mechanisms need to interoperate with one another, one of the major obstacles is the inability of the decryptor product to recognize, and optionally validate, the key recovery information generated by the encryptor product. In this paper, a common Key Recovery Block (KRB) format is being proposed to facilitate interoperability between heterogeneous key recovery systems. The KRB serves as a container for mechanism-specific key recovery information, and supports techniques to identify and optionally validate the contained key recovery information. This specification provides an extensible set of KRB validation techniques - however, it makes no attempt to set a preference for one technique over the others. The choice of validation technique(s) used is determined by the policies (with respect to the use of cryptography and key recovery) that apply to the encryptor and decryptor products. It may be noted that the KRB format specification is independent of the encryption algorithm used to protect the confidentiality of the data, and independent of the communication or storage protocol used to carry the encrypted data. It should also be recognized that the KRB format proposed in this paper is of limited scope. It assumes that the key recovery information can be made available to interested parties along with the encrypted data. There are a number of open issues regarding the techniques that allow key recovery information to be associated with encrypted data (whether the key recovery information and the encrypted data are transmitted over the same channel or separate channels) — these issues are beyond the scope of this proposal.

Fault-tolerant wormhole routing using a variation of the distributed recovery block approach

A fault-tolerant wormhole routing technique that incorporates a variation of the distributed recovery block (DRB) approach is described. The section of a parallel system that spans between the source and destination nodes is dynamically partitioned into overlapping DRB groups. A DRB group consists of a current node, a primary and an alternate successor node. The message packets travel towards the destination from one DRB group to the next group. A prototype of the routing system is implemented for mesh and hypercube topologies; however, the method can be used for topologies with a minimum node connectivity of three. The simulation results indicate that the DRB approach based wormhole routing tolerates both node and link failures.

Global Interoperability for Key Recovery

This document describes techniques for implementing key recovery. The document then addresses interoperability between encryptors and decryptors using the key encapsulation method of key recovery. The Common Key Recovery Block is presented as a cornerstone for interoperability. The interoperability is summarized in an interoperability matrix for key recovery systems.

Optimization models for recovery block schemes

This paper presents optimization models for a fault tolerant software by selecting a set of versions for a given program. The objective is to maximize the reliability of the software satisfying a budget limitation. Optimization models are developed for two block recovery schemes: (1) independent recovery block and (2) consensus recovery block. The paper also presents simple formulas to calculate the reliability of the two schemes.

Remarks on KRA key recovery block format

The introduction of a plethora of key recovery (KR) schemes and the lack of a standard has led to interoperability problems between dissimilar mechanisms. To overcome these problems the key recovery alliance (KRA) has proposed a common key recovery block (KRB) format. The authors identify some cases in which the mechanism fails to achieve its objectives.

Optimization models for complex recovery block schemes

In this paper we consider three complex recovery block schemes: (1) nested recovery block; (2) modified recovery block; and (3) nested consensus recovery block. For all the three models: (1) we derive simple expressions for calculating the system reliability; (2) we investigate how to arrange the versions in a nested recovery block and in a nested consensus recovery block; and (3) we develop optimization models where the objective is to optimize the reliability satisfying a budget constraint. The paper also includes efficient branch and bound procedures to solve the optimization problems. Two numerical example problems are solved to illustrate the branch and bound technique. Scope and purposes Software has become an integral part of many systems. For systems such as air traffic control, space shuttle, fighter aircraft and automated-guided missiles, it is usually the reliability of computer software that has the most single significant impact on the performance. Virtually, all major military systems are dependent on the correct operation of defense systems’ software. For systems such as air traffic control, fighter aircraft and automated guided missiles, 80% of the life cycle cost is spent on the software. Traditionally, reliability of any hardware and software is improved by introducing redundancy. However, redundant components require additional resources. Therefore, the redundancy level needed to achieve specified reliability must be carefully determined. The main purpose of this paper is to develop reliability prediction techniques and optimization models for important fault-tolerant software such as nested recovery block, modified recovery block and nested consensus recovery block. The methods have significant use in selection of programs in COTS (commercial-off-the-shelf) environment.

Reliability Analysis of Fault Tolerant Recovery Blocks

Reliability models for fault tolerant recovery blocks with exponential execution time are analysed. Explicit expressions for the reliability of independent recovery block, with exponential execution time are derived. The paper presents conditions for the optimal arrangement of versions within a recovery block. The paper also presents two optimisation models for the independent recovery blocks with exponential execution time. The objective is to maximise the reliability of the fault tolerant software satisfying a budget constraint and a constraint on the mean execution time of the software. Simple branch and bound procedures are developed that can be used to solve the optimisation problem.