Experimental evaluation of software aging effects in the eucalyptus elastic block storage

Matos, Rubens; Araújo, Jean; Alves, Vandi; Maciel, Paulo

doi:10.1109/icsmc.2012.6377878

Cited by 13 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Software rejuvenation is a technique that usually improves system availability [32]. Rejuvenation has the function of preventing crashes and reducing performance.…”

Section: Model C: Model B With Rejuvenation Technique Inclusionmentioning

confidence: 99%

“…The sensitivity analysis presented later pointed out the edge server as the most sensitive component. Thus, rejuvenation was applied to the edge server since many studies prove that servers suffer from software aging, which impacts system availability [32,34,35]. The rejuvenation model was based on the model proposed by Garg et al [36].…”

Section: Model C: Model B With Rejuvenation Technique Inclusionmentioning

confidence: 99%

See 1 more Smart Citation

Model-Driven Impact Quantification of Energy Resource Redundancy and Server Rejuvenation on the Dependability of Medical Sensor Networks in Smart Hospitals

Silva

Brito

Araújo

et al. 2022

Sensors

View full text Add to dashboard Cite

The spread of the Coronavirus (COVID-19) pandemic across countries all over the world urges governments to revolutionize the traditional medical hospitals/centers to provide sustainable and trustworthy medical services to patients under the pressure of the huge overload on the computing systems of wireless sensor networks (WSNs) for medical monitoring as well as treatment services of medical professionals. Uncertain malfunctions in any part of the medical computing infrastructure, from its power system in a remote area to the local computing systems at a smart hospital, can cause critical failures in medical monitoring services, which could lead to a fatal loss of human life in the worst case. Therefore, early design in the medical computing infrastructure’s power and computing systems needs to carefully consider the dependability characteristics, including the reliability and availability of the WSNs in smart hospitals under an uncertain outage of any part of the energy resources or failures of computing servers, especially due to software aging. In that regard, we propose reliability and availability models adopting stochastic Petri net (SPN) to quantify the impact of energy resources and server rejuvenation on the dependability of medical sensor networks. Three different availability models (A, B, and C) are developed in accordance with various operational configurations of a smart hospital’s computing infrastructure to assimilate the impact of energy resource redundancy and server rejuvenation techniques for high availability. Moreover, a comprehensive sensitivity analysis is performed to investigate the components that impose the greatest impact on the system availability. The analysis results indicate different impacts of the considered configurations on the WSN’s operational availability in smart hospitals, particularly 99.40%, 99.53%, and 99.64% for the configurations A, B, and C, respectively. This result highlights the difference of 21 h of downtime per year when comparing the worst with the best case. This study can help leverage the early design of smart hospitals considering its wireless medical sensor networks’ dependability in quality of service to cope with overloading medical services in world-wide virus pandemics.

show abstract

“…Software rejuvenation is a technique that usually improves system availability [32]. Rejuvenation has the function of preventing crashes and reducing performance.…”

Section: Model C: Model B With Rejuvenation Technique Inclusionmentioning

confidence: 99%

Section: Model C: Model B With Rejuvenation Technique Inclusionmentioning

confidence: 99%

Model-Driven Impact Quantification of Energy Resource Redundancy and Server Rejuvenation on the Dependability of Medical Sensor Networks in Smart Hospitals

Silva

Brito

Araújo

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Researchers studying system aging have hypothesized the close correlation between long-term usage of software and system performance degradation. Moreover, they studied the causes for system aging with an emphasis on long-term software usage [1,6,7,12], such as the exhaustion of the memory swap space of the operating system [9,13] and resource exhaustion [23,24] when using a software for a long time. Therefore, operating a software for long periods to establish experimental conditions and mimic actual system aging is rather difficult.…”

Section: Limitations Of the System-aging Related Researchmentioning

confidence: 99%

“…System aging is observed not only in personal computer/server-based computer systems but also in emerging computing systems such as cloud computing [2], mobile [3], and IoT enhanced with 5G network [4] systems, and there has been a significant amount of research analyzing the reasons for system aging [5]. To determine the aging within the system kernel range, various related studies have been carried out within the range of the system kernel [6][7][8][9][10][11] to the user application [12], from kernel-based research on system aging based on the development of kernel tracing tools, the collection of parameters related to several kernel subsystems, and a statistical analysis of the collected data, to research hypothesizing that the reason for system aging is an element of the software design and errors in the software itself. However, finding an accurate reason for system aging is still regarded as extremely difficult research [13].…”

Section: Introductionmentioning

confidence: 99%

Aging Analysis Framework of Windows-Based Systems through Differential-Analysis of System Snapshots

Jang¹,

Baek²,

Park³

2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

When a Windows-based system is used for an exceedingly long time, its performance degrades, and the error occurrence rate tends to increase. This is generally called system aging. To investigate the reasons for system aging, various studies have been conducted within the range of the operating system kernel to the user application. However, finding an accurate reason for system performance degradation remains challenging research topic. In this study, system monitoring was conducted by dividing a system into 'before software installation,' 'after software installation,' and 'after software removal.' We confirmed that when a software installed in a system is removed, various system elements, such as storage and memory, are not restored to the level prior to the software installation. Consequently, we established a hypothesis regarding the performance degradation of a computer system owing to repeated software installation/removal operations, investigated the correlation between system aging and repeated software installation/removal operations, and proposed a system aging analysis framework for analyzing the reason behind system aging. In the proposed system aging analysis framework, we aim to forcibly age a Windows-based system by repeating the software installation/removal operation by utilizing the system forced aging module. The framework identifies the elements affecting system performance through a differential data analysis of the system time-series data extracted by the system performance extraction and system component snapshot modules. Consequently, the aging analysis framework presented in this study is expected to be effectively utilized as an index for studying system aging.

show abstract

“…Alonso and Torres presented a monitoring framework based on aspect‐oriented programming to monitor the resource usage of every component in a web application, thus determining which component is related to software aging. To investigate the software aging affects in the cloud computing infrastructure, the virtual and resident memory usages were monitored; RAM memory exhaustion, swap memory depletion, high CPU utilization, and the subsequent increasing in the response time of applications were observed …”

Section: Introductionmentioning

confidence: 99%

Software Aging and Rejuvenation in a J2EE Application Server

Meng¹,

Hei²,

Zhang³

et al. 2014

Qual. Reliab. Engng. Int.

View full text Add to dashboard Cite

Recently, software aging and software rejuvenation have been widely researched by both academic and industrial communities. Software aging is characterized by progressive performance degradation or a sudden hang/crash in a longrunning software system due to resource exhaustion of the operating system. Software rejuvenation is a proactive and preventive maintenance technique to counteract software aging. In this work, the memory leak in Java virtual machine (JVM) is firstly analyzed, and then the software aging phenomenon in a J2EE application server is investigated. To mitigate software aging effects and prevent the system from failing, the hierarchical software rejuvenation method by partial or total restarting of the system is presented. Finally, the experimental results proved that the proposed method is superior to the periodical rejuvenation method in the aspect of improving system performance.

show abstract

Experimental evaluation of software aging effects in the eucalyptus elastic block storage

Cited by 13 publications

References 13 publications

Model-Driven Impact Quantification of Energy Resource Redundancy and Server Rejuvenation on the Dependability of Medical Sensor Networks in Smart Hospitals

Model-Driven Impact Quantification of Energy Resource Redundancy and Server Rejuvenation on the Dependability of Medical Sensor Networks in Smart Hospitals

Aging Analysis Framework of Windows-Based Systems through Differential-Analysis of System Snapshots

Software Aging and Rejuvenation in a J2EE Application Server

Contact Info

Product

Resources

About