Maximizing interval reliability in operational software system with rejuvenation

The reliability and availability of a cloud-based system play an important role in evaluating its system performance. Due to the promised high reliability of physical facilities provided for cloud services, software faults have become a major factor for failures of cloud-based systems. In this paper, we focus on the software aging phenomenon where system performance may be progressively degraded due to exhaustion of system resources, fragmentation and accumulation of errors. We present a proactive technique, called software rejuvenation, to counteract the software aging problem. The dynamic fault tree (DFT) formalism is adopted to model the system reliability before and during a software rejuvenation process in an aging cloud-based system. Then it is converted into Markov Chains to derive the system reliability function. We use a case study of a cloud-based system to illustrate the validity of our approach. Based on the reliability analysis results, we show how to estimate software rejuvenation schedules that can keep the system reliability above a predefined critical level for required system availability.

Android users are occasionally troubled by the slow UI responses and sudden application/OS crashes. These problems are mainly caused by software aging, a phenomenon of progressive degradation of performance and dependability typically observed in long-running software systems. A countermeasure to software aging is software rejuvenation, i.e., manual or scheduled restart at different levels, such as application, OS, and device. Various software aging and rejuvenation models have been proposed for different software systems. However, these traditional models cannot be applied in the context of mobile devices, as they seldom consider the patterns of usage behavior and user experience specific to mobile phones. We address this problem based on the observations that the usage time of mobile phones is typically fragmented in daily life, with frequent and periodical switches between active and sleep modes, and that the user experience on fluent operation in the active mode is a key concern for mobile users. These insights are exploited to model the usage behavior and aging process by individual Stochastic Petri-Nets, and then to compose them into a Continuous Time Markov Chain (CTMC). Furthermore, we propose proactive rejuvenation strategies based on such CTMCs to achieve the best user experience and the least user interference, such as restarting the device when it is in sleep mode and before it enters an aged state. To consider user experience - a key concern of mobile users which is still less prominent in traditional dependability measurements – we propose new related metrics: for fluency (i.e., the probability that a phone offers a fast UI response to the users), and for the degree of overall user experience. We demonstrate the effectiveness and advantages of the proposed models and metrics via simulations as well as an empirical study.

As businesses increasingly rely on IT for their mission-critical operations, continuous availability is a universal concern. Nowadays, virtualized platform has become the popular option to deploy complex enough services. Deployed services are expected to be always, available, but these long running services are especially sensitive to suffer from software aging phenomenon. Software aging of virtual machine monitors (VMMs) is becoming critical because performance degradation or crash failure of a VMM affects all virtual machines (VMs) on it. To counteract this issue, we deploy software rejuvenation methodology. To eliminate the service outage during software rejuvenation process, we combine the rejuvenation with live migration technology. Live VM migration enables a running VM on a host serer to move onto the other host server with very small interruption of the execution. Live VM migration depends on VMs placement and efficient resource available is required. The idea behind our paper is two-fold. First, we present the optimization of the resource usage as accepting as many services as in virtualized environment which support of VM live migration. Second, to demonstrate how much it can improve system availability, the stochastic Petri nets model a virtualized server system in case of using time based software rejuvenation for VMM is presented. Finally, we perform the numerical analysis to evaluate the model.

Software aging often affects the performance of a software system and eventually causes it to fail. To highlight the aging problems in long running applications, we have conducted the study of software rejuvenation approach and virtualization technology in this paper. In order to reduce the downtime during rejuvenation interval, software rejuvenation and virtualization technology are combined in this paper. Virtualization allows multiple operating system instances to run concurrently on a single physical machine. In this paper, we map virtualization based software rejuvenation approach with Markov Chain process and get mathematical steady-state solutions of the chain. We model multiple physical servers which are used to host multiple virtual machines (VMs) using stochastic modeling and evaluate it through both numerical analysis and SHARPE (Symbolic Hierarchical Automated Reliability and Performance Evaluator) tool simulation. The numerical results show that software rejuvenation and virtualization can be used to prolong the availability of the services and reduce the downtime of the systems.

Neural network based approach for time to crash prediction to cope with software aging

This paper present the interval reliability of a semi-Markov systems in general state space in continuous and discrete-time cases. We get also, as a particular case the interval reliability for the alternating renewal process. This is a continuation of the paper Limnios (Methodol Comput Appl Probab 14(4):895–917, 2012) where the interval reliability is referred as interval availability and also Propositions 2.1 and 3.1 below replace Propositions 3.4 and 5.3 respectively.

The phenomenon that the state of software degrades with time is known as software aging. The primary method to fight aging is software rejuvenation. This paper presents new ways of effective software rejuvenation using virtualization for addressing software aging. This new approach is meant to be the less disruptive as possible for the running service and to get a zero downtime in most of the cases. We construct the state transition models to describe the behaviors of virtualized and non- virtualized application server. We map through the rejuvenation actions to this transition model with stochastic process and express availability, downtime and downtime costs in terms of the parameters in our models. Our results show that virtualization and software rejuvenation can be used to prolong the availability of the services.

Continuously running software systems suffer from performance degradation due to aging phenomena and resource exhaustion that are mainly due to error conditions accumulation or due to exhaustion of system resources such as the physical memory. To counteract such phenomena, a periodic preemptive rollback to prevent software failures in the future, called software rejuvenation, can be applied. During the past 20 years, software rejuvenation has been extensively studied with the aim to design rejuvenation policies that optimize system availability, reliability, and performance, mainly in terms of operational cost. In this paper, we propose to model software systems’ overall performance capacity by assigning a performance capacity level at each of the possible states that it can be in, using a continuous time Markov process. A performance capacity indicator for all possible rejuvenation models incorporating partial, full, or both rejuvenation actions is defined and evaluated in the transient, and the steady state phase and the impact of various rejuvenation policies on it are further examined. Our aim is to distinguish the optimal, in terms of performance, rejuvenation strategy consisting of the type of rejuvenation to be implemented and the rejuvenation interval as well. Multi-objective optimization problems are additionally formulated for deriving the rejuvenation policies that optimize the system’s overall performance capacity with respect to availability and operational cost constraints. Numerical examples, in which through comparisons among the rejuvenation models and policies we can distinguish the optimal rejuvenation strategy to be followed, are also presented.

Since Component-Based Software Development technology has been widely used, traditional software rejuvenation strategy seems too rough on solving Component-Software aging, the whole system will be regenerated even only a few components aging, which leads to high cost of software rejuvenation and reduces the stability of the system. Therefore, we propose a component-based software rejuvenation strategy and we make components be the objects of rejuvenation. The aging components will be rejuvenated when software ages. In this paper, the process of judging software aging and software rejuvenation are described, then we propose the methods of judging software aging, determining the sequence that the aging components to be rejuvenated and rejuvenating components. Finally, an experiment is done to certify the efficiency of the strategy.

Software aging affects the availability of Virtual Machine Monitor (VMM), one of the main components of virtualized environments. Software aging causes internal software degradation due to bugs activation and accumulation during its execution. Software rejuvenation implemented through Virtual Machine (VM) Live Migration may be applied to cope with software aging effects on cloud computing system. Another relevant problem is the power consumption of using software rejuvenation techniques in the virtualized environments. This paper presents availability models based on Stochastic Petri Nets to evaluate two VM Live Migration approaches. These approaches are based on the redundancy schemes: Warm-Standby and Cold-Standby. In the Cold-Standby migration, the migration target machine is started only before the VM Live Migration. In the Warm-Standby Migration, the migration target machine runs along with the source migration machine. Results show that VM Live Migration causes a significant improvement in system availability. Scenarios with a heavy workload present an annual downtime reduction of 164 h. The availability comparison between two approaches reveals that the Cold-Standby approach has a slightly better result due to the decrease in the total number of VM Live Migrations. The power consumption results show that Cold-Standby approach is more efficient in power consumption. In all the observed scenarios, the costs savings by using Cold-Standby approach exceed 40%. The highlights of this paper are: i) a comprehensive model for availability evaluation of cloud computing with VMM rejuvenation through VM migration scheduling; ii) sensitivity analysis to define proper rejuvenation scheduling to maximize system availability and iii) power consumption comparison of the two adopted migration approaches.

A proactive approach towards always-on availability in broadband cable networks

Several studies have been conducted in order to understand the 'software aging' phenomenon. This paper presents the results of an experimental research work, which evaluated the software aging effects on a web server as well as the performance of a rejuvenation agent. Its main contribution is the identification and analytical validation of the factors that contribute for the web server aging. We adopted the design of experiment technique (DOE) to characterize the aging phenomenon. The results showed evidence that the 'page type' and 'page size' factors were responsible for over 99% of memory size variation in httpd processes. A 'rejuvenation agent' was implemented and integrated into the server environment to mitigate the aging effects. As a result, the size of the httpd processes could be kept under control and a higher availability and performance of the web server could be provided, as compared to the scenario without rejuvenation.

The problem of software aging widely exists in long-running software system, and the solution is software rejuvenation. Traditional software rejuvenation strategy has some deficiencies in solving the software aging problems of web services, for example, the higher failure rate and the poorer stability. Therefore, considering web services may have the correlation, at first we classified web service relationship and analyzed the relationship of web services by using the association mining algorithm; secondly we established and continuously fixed the aging model of web services by using the error back-propagation network to calculate the aging weights of web services. On this basis, based on the analysis of web services’ priority, call frequency and aging states, we calculated the rejuvenation cost, constructed the rejuvenation queue, and proposed an adaptive rejuvenation strategy of related web services. The experiment result showed that, compared with traditional regeneration strategy, this strategy improved the stability and dependability of web services.

Long running software systems are known to experience an aging phenomenon called software aging, one in which the accumulation of errors during the execution of software leads to performance degradation and eventually results in failure. To counteract this phenomenon an active fault management approach, called software rejuvenation, is particularly useful. It essentially involves gracefully terminating an application or a system and restarting it in a clean internal state. We deal with dependability analysis of a client/server software system with rejuvenation. Three dependability measures in the server process, steady-state availability, loss probability of requests and mean response time on tasks, are derived from the well-known hidden Markovian analysis under the time-based software rejuvenation scheme. In numerical examples, we investigate the sensitivity of some model parameters to the dependability measures.

Проведено огляд та аналіз літературних джерел, в яких досліджено явища старіння програмного забезпечення. Процес старіння охарактеризовано як погіршення продуктивності і збільшення кількості відмов, що має негативний вплив на показники надійності програмного забезпечення. Встановлено, що помилки програмного забезпечення та їх накопичення протягом виконання програми є причиною виникнення старіння програмного забезпечення. Визначено основні поняття та характеристики, що стосуються явища старіння, зокрема ефекти, чинники та метрики старіння, час до виснаження ресурсів, час до відмови старіння та робоче навантаження. Розглянуто класифікацію чинників старіння програмного забезпечення. Встановлено, що чинники можуть бути загальні для всіх систем і спеціальні для конкретних систем, зокрема мобільних. Здійснено порівняльний аналіз основних методів та підходів до моделювання процесу старіння програмного забезпечення. З'ясовано, що розроблення гібридних підходів та моделей, які включають переваги аналітичних моделей та моделей на основі вимірювань, є перспективним напрямом у вивченні проблеми старіння ПЗ. Показано, що мобільні операційні системи та додатки є особливо чутливими до ефектів старіння, оскільки вони працюють тривалий час без перезавантаження та часто мають обмежені ресурси, такі як пам'ять. Обґрунтовано актуальність урахування впливу цього явища для забезпечення надійності сучасних мобільних і вбудованих систем.