Prediction of Distributed Systems State Based on Monitoring Data

Drăghici, Adriana; Costan, Alexandru; Cristea, Valentin

doi:10.1109/ispdc.2010.28

Cited by 4 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This model also has the ability to provide confident level for each prediction it made and perform continuous adaptation. Draghisi A., A. Costan, V. Cristea [16], presented a presentation architecture developed within the MONALISA monitoring framework which also provides methods for estimating future values for different parameters on various periods of time. These predictions enhances the self adaptive behavior of several data intensive applications.…”

Section: Literature Studymentioning

confidence: 99%

Analyzing Process Behavior to Predict Resource Allocation in Distributed Environment by Using Time Series and Online Predictive Approach Algorithm

Galgali¹,

Vaidya²,

Tidke³

2014

IOSRJCE

View full text Add to dashboard Cite

A distributed system is a collection of different computers to handle large amount of data, the connected computers can share and coordinate their data on network. It is very difficult to for server to handle, analyze & process such a large amount of data, so that the performance of operations performed by these systems is reduced and they will produce inefficient data as a result. To handle this problem this paper has implemented the online predictive approach algorithm which uses time series to monitor process behavior, analyze it and predict the future observations to determine the resources required by the process in future.The resource prediction is used to optimize the data access operations like read, write, uploading and downloading the file to and from distributed system & which will improve the performance of distributed system. Time series is generated for every operation so that client will get performance chart.

show abstract

Section: Literature Studymentioning

confidence: 99%

Analyzing Process Behavior to Predict Resource Allocation in Distributed Environment by Using Time Series and Online Predictive Approach Algorithm

Galgali¹,

Vaidya²,

Tidke³

2014

IOSRJCE

View full text Add to dashboard Cite

show abstract

“…The techniques offered by the Prediction API are thoroughly described in [4], and consist of three predictors, filtering methods and a class for evaluation. The API can also be used independently from the MonALISA web interface, by providing as input data files in a specific format(ARFF) and obtaining as output images with plots and files with predicted values.…”

Section: Prediction Algorithms Overviewmentioning

confidence: 99%

“…The algorithm uses the training data to built the regression line on different intervals(also referred as windows) and evaluate the errors between the real and the predicted values. Details regarding the mathematics foundations of regression line can be found in [4]. As a data preprocessing technique, Outliers elimination is essential for the prediction techniques based on linear regression.…”

Section: Prediction Algorithms Overviewmentioning

confidence: 99%

Prediction Strategies for Self-Adaptive Behavior in Distributed Systems

Costan

Drăghici

Cristea

2010

2010 International Conference on P2P, Parallel, Grid, Cloud and Internet Computing

Self Cite

View full text Add to dashboard Cite

The autonomic management of large-scale distributed systems now allows performance improvement, availability, and security, while simultaneously reducing the effort and skills required of system administrators. One way that systems can support these abilities is by relying on a continuous monitoring service to keep track of the states of the targeted systems. However, it is challenging to achieve both scalability and high accuracy when dealing with huge amounts of distributed and time-varying metrics in large-scale production systems. Additionally, most of the monitoring services are designed to provide general resource information and do not consider specific information for higher-level services, lacking important control capabilities. In this context, a dynamic adaptation layer is required, to be able to reactive and proactive deal with detected or predicted conditions. In this paper, we present a prediction architecture developed within the MonALISA monitoring framework, providing methods for estimating future values for different parameters on various periods of time. The predictions are used to enhance the self-adaptive behavior of applications and to minimize continuous monitoring costs. We illustrate our work as an efficient strategy to save power in wireless sensor networks by reducing the data traffic and thus maximizing the network lifetime.

show abstract

“…Monitoring provides information on the program's runtime behavior that cannot be easily obtained from the program's source code alone. Such information is very useful for testing and debugging [4,5].…”

Section: Introductionmentioning

confidence: 99%

A tool for monitoring and tracing multi-agent applications

Maghayreh

Doush

Alkhateeb

et al. 2011

2011 24th Canadian Conference on Electrical and Computer Engineering(CCECE)

View full text Add to dashboard Cite

In this paper, we have designed and implemented a tool that can be used to effectively collect the trace data of multi-agent applications developed using the Java Agent DEvelopment Framework (JADE). The tool can also be used to monitor JADE applications and to report the occurrence of certain events of interest. The collected trace data can be used later by the developer to verify that the current implementation of the multi-agent application under consideration satisfies certain properties. Moreover, monitoring the application at runtime and reporting the occurrence of certain events can significantly help the developers in testing and debugging multiagent applications.

show abstract

Prediction of Distributed Systems State Based on Monitoring Data

Cited by 4 publications

References 8 publications

Analyzing Process Behavior to Predict Resource Allocation in Distributed Environment by Using Time Series and Online Predictive Approach Algorithm

Analyzing Process Behavior to Predict Resource Allocation in Distributed Environment by Using Time Series and Online Predictive Approach Algorithm

Prediction Strategies for Self-Adaptive Behavior in Distributed Systems

A tool for monitoring and tracing multi-agent applications

Contact Info

Product

Resources

About