Operating System Performance

A software aging in convoluted system refers to the situation where software degrades with span of time. This phenomenon, which may eventually lead to system performance degradation or crash/hang failure, is the result of depletion of operating system resources, data deception and numerical error assembly. A technique called software rejuvenation has been incorporated, which essentially involves periodic aborting an application or a system, flushing its intramural state and re-starting it. A main issue in rejuvenation is to discover ideal time to initiate software rejuvenation. Software rejuvenation is a proactive technique that allows preventing the occurrence of software failing. A novel approach called Smart interval and payload (SIP) policy is introduced to overcome all the hurdles in the present scenario based on Software Rejuvenation approaches. SIP policy accepts time from user and optimizes the rejuvenation time whenever workload is variable; otherwise the system is rejuvenated at its rejuvenation point. SIP policy avoids software failure and it helps to achieve high availability of convoluted system.

Download Full-text

Performance Evaluation for Dynamic Voltage and Frequency Scaling Using Runtime Performance Counters

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.284-287.2575 ◽

2013 ◽

Vol 284-287 ◽

pp. 2575-2579 ◽

Cited By ~ 1

Author(s):

Wen Yew Liang ◽

Ming Feng Chang ◽

Yen Lin Chen ◽

Jenq Haur Wang

Keyword(s):

Operating System ◽

Performance Evaluation ◽

Execution Time ◽

System Performance ◽

Evaluation Method ◽

Reducing Power ◽

General Purpose ◽

Frequency Scaling ◽

Dynamic Voltage ◽

Performance Evaluation Method

Dynamic voltage and frequency scaling (DVFS) is an effective technique for reducing power consumption. The system performance is not easy to evaluate through Dynamic Voltage and Frequency Scaling. Most of studies use the execution time as an indicator while measuring the performance. However, DVFS adjusted processor speed during a fixed-length period so it cannot rely on the execution time to evaluate the system performance. This study proposes a novel and simple performance evaluation method to evaluate the system performance when DVFS is activated. Based on the performance evaluation method, this study also proposes a DVFS algorithm (P-DVFS) for a general-purpose operating system. The algorithm has been implemented on the Linux operating system and used a PXA270 development board. The results show that P-DVFS could accurately predict the suitable frequency, given runtime statistics information of a running program. In this way, the user can easily control the energy consumption by specifying allowable performance loss factor.

Download Full-text

Operating system performance measurements for Remote Patient Monitoring applications

2013 IEEE International Symposium on Medical Measurements and Applications (MeMeA) ◽

10.1109/memea.2013.6549765 ◽

2013 ◽

Cited By ~ 5

Author(s):

V. Joshi ◽

P. Moradshahi ◽

R. Goubran

Keyword(s):

Operating System ◽

System Performance ◽

Patient Monitoring ◽

Remote Patient Monitoring ◽

Performance Measurements ◽

Monitoring Applications ◽

Remote Patient

Download Full-text

The impact of architectural trends on operating system performance

Proceedings of the fifteenth ACM symposium on Operating systems principles - SOSP '95 ◽

10.1145/224056.224078 ◽

1995 ◽

Cited By ~ 87

Author(s):

M. Rosenblum ◽

E. Bugnion ◽

S. A. Herrod ◽

E. Witchel ◽

A. Gupta

Keyword(s):

Operating System ◽

System Performance ◽

The Impact

Download Full-text

Autenticação Contínua de Usuários Utilizando Contadores de Desempenho do Sistema Operacional

10.5753/sbseg.2019.13979 ◽

2019 ◽

Author(s):

César Andrade ◽

Paulo Henrique Gonçalves ◽

Hendrio Bragança ◽

Eduardo Souto

Keyword(s):

Operating System ◽

Performance Indicators ◽

System Performance ◽

Data Classification ◽

Behavioral Model ◽

Assessment Methodology ◽

Behavioral Models ◽

Source Data ◽

Using Data

Computer authentication systems based on login and password have been vulnerable to the action of unauthorized users. Currently, authentication techniques based on behavioral models predominantly use information extracted from mouse and/or keyboard to authenticate users. Operating system performance indicators can be used as an alternative. This work proposes an approach using data from performance indicators such as source data, CNN/LSTM networks for data classification, and reliability-based assessment methodology for the purpose of authenticating the user on an ongoing basis. The results obtained demonstrate the feasibility of using these attributes as the origin of the data to define a behavioral model. The best result obtained in this research is that 100% of genuine users are never inadvertently blocked and 100% of the imposters are detected after the average of three actions. Sistemas de autenticação de computadores baseados em credencias de contas (e.g. login e senha) têm sido vulneráveis à ação de usuários não autorizados. Atualmente, as técnicas de autenticação baseadas em modelos comportamentais predominantemente usam informações extraídas de mouse e/ou teclado para autenticar os usuários. Contadores de desempenho de sistema operacional podem ser utilizadas como alternativa. Este trabalho propõe uma abordagem utilizando dados de contadores de desempenho como dados de origem, redes CNN/LSTM para classificação dos dados e metodologia de avaliação baseada em nível de confiança com o propósito de autenticar o usuário de forma contínua. Os resultados obtidos demonstram a viabilidade do uso destes atributos como origem dos dados para definição de modelo comportamental. O melhor resultado obtido nesta pesquisa é que 100% dos usuários genuínos nunca são bloqueados inadvertidamente e 100% dos impostores são detectados após a média de três ações.

Download Full-text