A Comprehensive Availability Modeling and Analysis of a Virtualized Servers System Using Stochastic Reward Nets

Abstract

It is important to assess availability of virtualized systems in IT business infrastructures. Previous work on availability modeling and analysis of the virtualized systems used a simplified configuration and assumption in which only one virtual machine (VM) runs on a virtual machine monitor (VMM) hosted on a physical server. In this paper, we show a comprehensive availability model using stochastic reward nets (SRN). The model takes into account (i) the detailed failures and recovery behaviors of multiple VMs, (ii) various other failure modes and corresponding recovery behaviors (e.g., hardware faults, failure and recovery due to Mandelbugs and aging-related bugs), and (iii) dependency between different subcomponents (e.g., between physical host failure and VMM, etc.) in a virtualized servers system. We also show numerical analysis on steady state availability, downtime in hours per year, transaction loss, and sensitivity analysis. This model provides a new finding on how to increase system availability by combining both software rejuvenations at VM and VMM in a wise manner.

A Hierarchical Modeling and Analysis Framework for Availability and Security Quantification of IoT Infrastructures

“…The proposed hierarchical models of the IoT smart factory infrastructure are all implemented in symbolic hierarchical automated reliability and performance evaluator (SHARPE) [58,59]. The input parameters are mostly based on previous experimental studies and consolidated works [19,27,[60][61][62] as shown (i) in Table 1 for default input parameters used in software/hardware sub-models of cloud member system, (ii) in Table 2 for default input parameters of software/hardware sub-models of fog member system, and (iii) in Table 3 for default input parameters of edge member system's software/hardware sub-models. The developed hierarchical model of the IoT smart factory infrastructure is analyzed in regard to various analysis outputs including (i) SSA, (ii) sensitivity of SSA wrt.…”

“…State-of-the-art computing paradigms: Cloud computing has been accredited as a centralized computing paradigm successfully adopted in many online business services and applications in the past few years featured by its pricing models of pay-as-you-go and its service platforms of Software as a Service (SaaS), Platform as a Service (PaaS) and Infrastructure as a Service (IaaS) [24,25]. Resource virtualization technologies along with fault-tolerance techniques at all levels of the cloud infrastructure help ease centralized operative management and resource scalability, thus maximize service availability, agility and adaptability to the non-uniform variation of service loads and user demands [26][27][28]. Nevertheless, the geographical centralization and the limited amount of huge cloud data centers which are often allocated in specific regions around the world for safety reasons cause various issues in real-time data transactions of latency-sensitive services and applications running in remote areas.…”

A Comprehensive Sensitivity Analysis of a Data Center Network with Server Virtualization for Business Continuity

“…Sensitivity analysis [10,18,37,38] is used popularly to provide a selection basis and help design system parameters by observing system characteristics and responses with respect to predetermined valuables in order to identify the most impacting factors as well as to detect bottlenecks in system availability. One may adopt two types of sensitivity analysis: (i) nonparametric sensitivity analysis [39], which studies the system responses upon the component addition/removal or modifications of system model and (ii) parametric sensitivity analysis [40], which observes the system behaviors with respect to the variations of given input parameters.…”

“…The DCell network architecture allows avoiding any single point of failure and thus is able to tolerate different types of failures such as node failures, link failures, and network device failures. Furthermore, to enhance system availability and capability of fault tolerance, one may employ server virtualization [7,[9][10][11] into a DCN. The approach creates virtual computing machines (VMs) on each physical host of the DCN.…”

Modeling and Evaluation of Power-Aware Software Rejuvenation in Cloud Systems

“…The tendency of software to fail or cause a system failure after running continuously for a specific time period is referred to as software aging [6,7]. Software aging is a phenomenon in long-run software systems that causes an increased failure rate and/or degraded performance due to accumulation of aging errors [8,9]. Since VM requests in IaaS clouds are usually different in terms of software and tools for which they are initiated, their corresponding VMs exhibit complex behaviors and sophisticated interactions throughout their lifetime that enable VMMs to manage a wide variety of VM behaviors.…”