Software aging and multifractality of memory resources

  title={Software aging and multifractality of memory resources},
  author={Mark Shereshevsky and John I. Crowell and Bojan Cukic and Vijai Gandikota and Yan Liu},
  journal={2003 International Conference on Dependable Systems and Networks, 2003. Proceedings.},
We investigate the dynamics of monitored memory resource utilizations in an operating system under stress using quantitative methods of fractal analysis. In the experiments, we recorded the time series representing various memory related parameters of the operating system. We observed that parameters demonstrate clear multifractal behavior. The degree of fractality of these time series tends to increase as the system workload increases. We conjecture that the H¨ older exponent that measures the… 

Figures from this paper

Modeling operating system crash behavior through multifractal analysis, long range dependence and mining of memory usage patterns
Three different techniques from fractal geometry are studied using the same datasets for operating system crash modeling and prediction, and all three approaches demonstrated apromising ability to identify software aging and predict upcoming operating system crashes.
Multi-scale Entropy: One Metric of Software Aging
This work extends classical multi-scale entropy analysis, proposes a novel aging metric called multidimensional multiscale entropy and proves that it have all the three properties of software aging through theory and experiment analysis.
A Nonlinear Approach to Modeling of Software Aging in a Web Server
A nonlinear dynamic model of software aging is constructed and shown to be nonlinear and chaotic to formulate software aging process of a widely used web server software system - the Apache HTTP server.
A Threshold Autoregressive Model for Software Aging
It is argued that software aging process could be divided into four stages: robust stage, transition stage, failure-probable stage and failure stage, and a non-linear threshold autoregressive (TAR) model is proposed to model and forecast the resource usage in these stages.
A Nonlinear Dynamic Model for Software Aging in Service-Oriented Software
This model can be used to forecast abrupt state degradation and help to explore the root cause of software aging, by comparing the output of the model against real values, with a suspected “aging factor” as input.
An approach for identifying failure-prone state of computer system
The memory thrashing phenomenon observed in a controlled experiment is reported, and the vibration frequency of available memory may be a potential indicator of aging, and a metric is proposed to measure the aging extent implicated in the vibration Frequency by using power spectrum analysis.
The mechanics of memory-related software aging
This paper goes deep into how memory management works inside application process, focusing on two memory problems that cause software aging: fragmenting and leakage, and explains the mechanics of memory-related software aging effects dissecting a real and widely adopted memory allocator.
A workload-based analysis of software aging, and rejuvenation
With the degradation model, it is proved that the failure rate is asymptotically constant in the absence of resource leaks, and it is increasing as leaks occur & accumulate, which confirms the resource leaks as an aging source.
CHAOS: Accurate and Realtime Detection of Aging-Oriented Failure Using Entropy
A novel entropy-based aging indicator, Multidimensional Multi-scale Entropy (MMSE), which employs the complexity embedded in runtime performance metrics to indicate software aging and leverages multi-scale and multi-dimension integration to tolerate system fluctuations.
Analysis of Software Aging in a Web Server
Based on the models employed here, proactive management techniques like software rejuvenation triggered by actual measurements can be built and how the exploitation of the seasonal variation can help in adequately predicting the future resource usage is shown.


A methodology for detection and estimation of software aging
A distributed data collection tool used to collect operating system resource usage and system activity data at regular intervals, from networked UNIX workstations and proposes a metric: "estimated time to exhaustion", which is calculated using well known slope estimation techniques.
A measurement-based model for estimation of resource exhaustion in operational software systems
A measurement-based model is proposed to estimate the rate of exhaustion of operating system resources both as a function of time and the system workload state, and a semi-Markov reward model is constructed based on workload and resource usage data collected from the UNIX operating system.
On the self-similar nature of Ethernet traffic
We demonstrate that Ethernet local area network (LAN) traffic is statistically self-similar, that none of the commonly used traffic models is able to capture this fractal behavior, and that such
On the self-similar nature of Ethernet traffic (extended version)
Demonstrates that Ethernet LAN traffic is statistically self-similar, that none of the commonly used traffic models is able to capture this fractal-like behavior, that such behavior has serious
Age-related alterations in the fractal scaling of cardiac interbeat interval dynamics.
The age-related loss of fractal organization in heartbeat dynamics may reflect the degradation of integrated physiological regulatory systems and may impair an individual's ability to adapt to stress.
Proactive management of software aging
This group has developed techniques that detect the occurrence of software aging due to resource exhaustion, estimate the time remaining until the exhaustion reaches a critical level, and automatically perform proactive software rejuvenation of an application, process group, or entire operating system.
Large Deviations and the Distribution of Price Changes
The Multifractal Model of Asset Returns ("MMAR," see Mandelbrot, Fisher, and Calvet, 1997) proposes a class of multifractal processes for the modelling of financial returns. In that paper,
Statistical non-parametric algorithms to estimate the optimal software rejuvenation schedule
The software rejuvenation models are formulated via the semi-Markov processes, and the optimalSoftware rejuvenation schedules which maximize the system availabilities are derived analytically for respective cases.
Detection of abrupt changes: theory and application
A unified framework for the design and the performance analysis of the algorithms for solving change detection problems and links with the analytical redundancy approach to fault detection in linear systems are established.
Revealing local variability properties of human heartbeat intervals with the local effective Hölder exponent
A moving average filtering of H\"older exponent based variability estimates was used to enhance these fluctuations and it is found that this way of local presentation of scaling properties may be of clinical importance.