S-SETA: Selective Software-Only Error-Detection Technique Using Assertions

Software-based techniques offer several advantages to increase the reliability of processor-based systems at very low cost, but they cause performance degradation and an increase of the code size. To meet constraints in performance and memory, we propose SETA, a new control-flow software-only technique that uses assertions to detect errors affecting the program flow. SETA is an independent technique, but it was conceived to work together with previously proposed data-flow techniques that aim at reducing performance and memory overheads. Thus, SETA is combined with such data-flow techniques and submitted to a fault injection campaign. Simulation and neutron induced SEE tests show high fault coverage at performance and memory overheads inferior to the state-of-the-art.

Error detection and correction in binary parallel digital computers /

"Retyped October, 1964"

Optimization of quantum states for signaling across an arbitrary qubit noise channel with minimum-error detection

International audience

Fault Tolerant Error Coding and Detection using Reversible Gates

In recent years, reversible logic has emerged as one of the most important approaches for power optimization with its application in low power CMOS, quantum computing and nanotechnology. Low power circuits implemented using reversible logic that provides single error correction – double error detection (SEC-DED) is proposed in this paper. The design is done using a new 4 x 4 reversible gate called ‘HCG’ for implementing hamming error coding and detection circuits. A parity preserving HCG (PPHCG) that preserves the input parity at the output bits is used for achieving fault tolerance for the hamming error coding and detection circuits.

Thresholds for error probability measures of business process models

The quality of conceptual business process models is highly relevant for the design of corresponding information systems. In particular, a precise measurement of model characteristics can be beneficial from a business perspective, helping to save costs thanks to early error detection. This is just as true from a software engineering point of view. In this latter case, models facilitate stakeholder communication and software system design. Research has investigated several proposals as regards measures for business process models, from a rather correlational perspective. This is helpful for understanding, for example size and complexity as general driving forces of error probability. Yet, design decisions usually have to build on thresholds, which can reliably indicate that a certain counter-action has to be taken. This cannot be achieved only by providing measures; it requires a systematic identification of effective and meaningful thresholds. In this paper, we derive thresholds for a set of structural measures for predicting errors in conceptual process models. To this end, we use a collection of 2,000 business process models from practice as a means of determining thresholds, applying an adaptation of the ROC curves method. Furthermore, an extensive validation of the derived thresholds was conducted by using 429 EPC models from an Australian financial institution. Finally, significant thresholds were adapted to refine existing modeling guidelines in a quantitative way.

Web应用服务器的适应性失效检测

失效检测是分布式系统的基本可靠性保障技术，它对运行时系统的存活状态进行及时检测．作为网络分布计算环境中的主流中间件，Web应用服务器（Web application server简称WAS）需要提供良好的检测机制，并且要能满足适应性的需求．适应性失效检测要求失效检测器能够根据应用需求和系统环境的变化而动态地改变检测的质量．首先给出了WAS的多层失效检测模型，然后基于失效检测器的服务质量规约，提出了适应性失效检测算法，并设计了一个WAS的适应性失效检测框架．它能够满足动态调整失效检测质量和灵活集成失效检测器的要求．该工作在OnceAS应用服务器中进行了实现，并给出了OnceAS平台上的实验及数据．

Heterogeneous error-resilient scheme for spectral analysis in ultra-low power wearable electrocardiogram devices

Wearable devices performing advanced bio-signal analysis algorithms are aimed to foster a revolution in healthcare provision of chronic cardiac diseases. In this context, energy efficiency is of paramount importance, as long-term monitoring must be ensured while relying on a tiny power source. Operating at a scaled supply voltage, just above the threshold voltage, effectively helps in saving substantial energy, but it makes circuits, and especially memories, more prone to errors, threatening the correct execution of algorithms. The use of error detection and correction codes may help to protect the entire memory content, however it incurs in large area and energy overheads which may not be compatible with the tight energy budgets of wearable systems. To cope with this challenge, in this paper we propose to limit the overhead of traditional schemes by selectively detecting and correcting errors only in data highly impacting the end-to-end quality of service of ultra-low power wearable electrocardiogram (ECG) devices. This partition adopts the protection of either significant words or significant bits of each data element, according to the application characteristics (statistical properties of the data in the application buffers), and its impact in determining the output. The proposed heterogeneous error protection scheme in real ECG signals allows substantial energy savings (11% in wearable devices) compared to state-of-the-art approaches, like ECC, in which the whole memory is protected against errors. At the same time, it also results in negligible output quality degradation in the evaluated power spectrum analysis application of ECG signals.

A block coprocessor for user data rate improvements to GPRS coding scheme 4

The general packet radio service (GPRS) has been developed to allow packet data to be transported efficiently over an existing circuit-switched radio network, such as GSM. The main application of GPRS are in transporting Internet protocol (IP) datagrams from web servers (for telemetry or for mobile Internet browsers). Four GPRS baseband coding schemes are defined to offer a trade-off in requested data rates versus propagation channel conditions. However, data rates in the order of > 100 kbits/s are only achievable if the simplest coding scheme is used (CS-4) which offers little error detection and correction (EDC) (requiring excellent SNR) and the receiver hardware is capable of full duplex which is not currently available in the consumer market. A simple EDC scheme to improve the GPRS block error rate (BLER) performance is presented, particularly for CS-4, however gains in other coding schemes are seen. For every GPRS radio block that is corrected by the EDC scheme, the block does not need to be retransmitted releasing bandwidth in the channel and improving the user's application data rate. As GPRS requires intensive processing in the baseband, a viable field programmable gate array (FPGA) solution is presented in this paper.

Contribution to flexible service coordination in mobile prosumer environments

Este trabajo de Tesis se desarrolla en el marco de los escenarios de ejecución distribuida de servicios móviles y contribuye a la definición y desarrollo del concepto de usuario prosumer. El usuario prosumer se caracteriza por utilizar su teléfono móvil para crear, proveer y ejecutar servicios. Este nuevo modelo de usuario contribuye al avance de la sociedad de la información, ya que el usuario prosumer se transforma de creador de contenidos a creador de servicios (estos últimos formados por contenidos y la lógica para acceder a ellos, procesarlos y representarlos). El objetivo general de este trabajo de Tesis es la provisión de un modelo de creación, distribución y ejecución de servicios para entorno móvil que permita a los usuarios no programadores (usuarios prosumer), pero expertos en un determinado dominio, crear y ejecutar sus propias aplicaciones y servicios. Para ello se definen, desarrollan e implementan metodologías, procesos, algoritmos y mecanismos adaptables a dominios específicos, para construir entornos de ejecución distribuida de servicios móviles para usuarios prosumer. La provisión de herramientas de creación adaptadas a usuarios no expertos es una tendencia actual que está siendo desarrollada en distintos trabajos de investigación. Sin embargo, no se ha propuesto una metodología de desarrollo de servicios que involucre al usuario prosumer en el proceso de diseño, desarrollo, implementación y validación de servicios. Este trabajo de Tesis realiza un estudio de las metodologías y tecnologías más innovadoras relacionadas con la co‐creación y utiliza este análisis para definir y validar una metodología que habilita al usuario para ser el responsable de la creación de servicios finales. Siendo los entornos móviles prosumer (mobile prosumer environments) una particularización de los entornos de ejecución distribuida de servicios móviles, en este trabajo se tesis se investiga en técnicas de adaptación, distribución, coordinación de servicios y acceso a recursos identificando como requisitos las problemáticas de este tipo de entornos y las características de los usuarios que participan en los mismos. Se contribuye a la adaptación de servicios definiendo un modelo de variabilidad que soporte la interdependencia entre las decisiones de personalización de los usuarios, incorporando mecanismos de guiado y detección de errores. La distribución de servicios se implementa utilizando técnicas de descomposición en árbol SPQR, cuantificando el impacto de separar cualquier servicio en distintos dominios. Considerando el plano de comunicaciones para la coordinación en la ejecución de servicios distribuidos hemos identificado varias problemáticas, como las pérdidas de enlace, conexiones, desconexiones y descubrimiento de participantes, que resolvemos utilizando técnicas de diseminación basadas en publicación subscripción y algoritmos Gossip. Para lograr una ejecución flexible de servicios distribuidos en entorno móvil, soportamos la adaptación a cambios en la disponibilidad de los recursos, proporcionando una infraestructura de comunicaciones para el acceso uniforme y eficiente a recursos. Se han realizado validaciones experimentales para evaluar la viabilidad de las soluciones propuestas, definiendo escenarios de aplicación relevantes (el nuevo universo inteligente, prosumerización de servicios en entornos hospitalarios y emergencias en la web de la cosas). Abstract This Thesis work is developed in the framework of distributed execution of mobile services and contributes to the definition and development of the concept of prosumer user. The prosumer user is characterized by using his mobile phone to create, provide and execute services. This new user model contributes to the advancement of the information society, as the prosumer is transformed from producer of content, to producer of services (consisting of content and logic to access them, process them and represent them). The overall goal of this Thesis work is to provide a model for creation, distribution and execution of services for the mobile environment that enables non‐programmers (prosumer users), but experts in a given domain, to create and execute their own applications and services. For this purpose I define, develop and implement methodologies, processes, algorithms and mechanisms, adapted to specific domains, to build distributed environments for the execution of mobile services for prosumer users. The provision of creation tools adapted to non‐expert users is a current trend that is being developed in different research works. However, it has not been proposed a service development methodology involving the prosumer user in the process of design, development, implementation and validation of services. This thesis work studies innovative methodologies and technologies related to the co‐creation and relies on this analysis to define and validate a methodological approach that enables the user to be responsible for creating final services. Being mobile prosumer environments a specific case of environments for distributed execution of mobile services, this Thesis work researches in service adaptation, distribution, coordination and resource access techniques, and identifies as requirements the challenges of such environments and characteristics of the participating users. I contribute to service adaptation by defining a variability model that supports the dependency of user personalization decisions, incorporating guiding and error detection mechanisms. Service distribution is implemented by using decomposition techniques based on SPQR trees, quantifying the impact of separating any service in different domains. Considering the communication level for the coordination of distributed service executions I have identified several problems, such as link losses, connections, disconnections and discovery of participants, which I solve using dissemination techniques based on publish‐subscribe communication models and Gossip algorithms. To achieve a flexible distributed service execution in mobile environments, I support adaptation to changes in the availability of resources, while providing a communication infrastructure for the uniform and efficient access to resources. Experimental validations have been conducted to assess the feasibility of the proposed solutions, defining relevant application scenarios (the new intelligent universe, service prosumerization in hospitals and emergency situations in the web of things).

Experience report: anomaly detection of cloud application operations using log and cloud metric correlation analysis

Failure of application operations is one of the maincauses of system-wide outages in cloud environments. Thisparticularly applies to DevOps operations, such as backup,redeployment, upgrade, customized scaling, and migration that areexposed to frequent interference from other concurrent operations,configuration changes, and resources failure. However, currentpractices fail to provide a reliable assurance of correct execution ofthese kinds of operations. In this paper, we present an approach toaddress this problem that adopts a regression-based analysistechnique to find the correlation between an operation’s activity logsand the operation activity’s effect on cloud resources. Thecorrelation model is then used to derive assertion specifications,which can be used for runtime verification of running operations andtheir impact on resources. We evaluated our proposed approach onAmazon EC2 with 22 rounds of rolling upgrade operations whileother types of operations were running and random faults wereinjected. Our experiment shows that our approach successfullymanaged to raise alarms for 115 random injected faults, with aprecision of 92.3%.

Impaired conflict monitoring in Parkinson's disease patients during an oculomotor redirect task

Fallibility is inherent in human cognition and so a system that will monitor performance is indispensable. While behavioral evidence for such a system derives from the finding that subjects slow down after trials that are likely to produce errors, the neural and behavioral characterization that enables such control is incomplete. Here, we report a specific role for dopamine/basal ganglia in response conflict by accessing deficits in performance monitoring in patients with Parkinson's disease. To characterize such a deficit, we used a modification of the oculomotor countermanding task to show that slowing down of responses that generate robust response conflict, and not post-error per se, is deficient in Parkinson's disease patients. Poor performance adjustment could be either due to impaired ability to slow RT subsequent to conflicts or due to impaired response conflict recognition. If the latter hypothesis was true, then PD subjects should show evidence of impaired error detection/correction, which was found to be the case. These results make a strong case for impaired performance monitoring in Parkinson's patients.

Enabling node repair in any erasure code for distributed storage

Erasure codes are an efficient means of storing data across a network in comparison to data replication, as they tend to reduce the amount of data stored in the network and offer increased resilience in the presence of node failures. The codes perform poorly though, when repair of a failed node is called for, as they typically require the entire file to be downloaded to repair a failed node. A new class of erasure codes, termed as regenerating codes were recently introduced, that do much better in this respect. However, given the variety of efficient erasure codes available in the literature, there is considerable interest in the construction of coding schemes that would enable traditional erasure codes to be used, while retaining the feature that only a fraction of the data need be downloaded for node repair. In this paper, we present a simple, yet powerful, framework that does precisely this. Under this framework, the nodes are partitioned into two types and encoded using two codes in a manner that reduces the problem of node-repair to that of erasure-decoding of the constituent codes. Depending upon the choice of the two codes, the framework can be used to avail one or more of the following advantages: simultaneous minimization of storage space and repair-bandwidth, low complexity of operation, fewer disk reads at helper nodes during repair, and error detection and correction.

Enabling node repair in any erasure code for distributed storage

Erasure codes are an efficient means of storing data across a network in comparison to data replication, as they tend to reduce the amount of data stored in the network and offer increased resilience in the presence of node failures. The codes perform poorly though, when repair of a failed node is called for, as they typically require the entire file to be downloaded to repair a failed node. A new class of erasure codes, termed as regenerating codes were recently introduced, that do much better in this respect. However, given the variety of efficient erasure codes available in the literature, there is considerable interest in the construction of coding schemes that would enable traditional erasure codes to be used, while retaining the feature that only a fraction of the data need be downloaded for node repair. In this paper, we present a simple, yet powerful, framework that does precisely this. Under this framework, the nodes are partitioned into two types and encoded using two codes in a manner that reduces the problem of node-repair to that of erasure-decoding of the constituent codes. Depending upon the choice of the two codes, the framework can be used to avail one or more of the following advantages: simultaneous minimization of storage space and repair-bandwidth, low complexity of operation, fewer disk reads at helper nodes during repair, and error detection and correction.

Uma análise dos esquemas de dígitos verificadores usados no Brasil

Neste trabalho discutimos vários sistemas de dígitos verificadores utilizados no Brasil, muitos deles semelhantes a esquemas usados mundialmente, e fazemos uma análise da sua capacidade de detectar os diversos tipos de erros que são comuns na entrada de dados em sistemas computacionais. A análise nos mostra que os esquemas escolhidos constituem decisões subotimizadas e quase nunca obtêm a melhor taxa de detecção de erros possível. Os sistemas de dígitos verificadores são baseados em três teorias da álgebra: aritmética modular, teoria de grupos e quasigrupos. Para os sistemas baseados em aritmética modular, apresentamos várias melhorias que podem ser introduzidas. Desenvolvemos um novo esquema ótimo baseado em aritmética modular base 10 com três permutações para identificadores de tamanho maior do que sete. Descrevemos também o esquema Verhoeff, já antigo, mas pouquíssimo utilizado e que também é uma alternativa de melhoria para identificadores de tamanho até sete. Desenvolvemos ainda, esquemas ótimos para qualquer base modular prima que detectam todos os tipos de erros considerados. A dissertação faz uso ainda de elementos da estatística, no estudo das probabilidades de detecção de erros e de algoritmos, na obtenção de esquemas ótimos.