Mathematical Model of Single-Channel Infocommunication Node with Several Packet Flows

Pustovoitov, Pavel; Sokol, Galina; Hroza, Petro; Tyrtyshnikov, Oleksii; Rvachova, Natalia

doi:10.1109/infocommst.2018.8632128

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This task can be solved, for example, using queuing theory [3] under the assumption that the intensity of the incoming and outgoing query flows [4] are known. The intensity of queries can be estimated using statistics from server but their intensities very in day time and in days of the week [5][6]. So, the task of the paper is to develop the mathematical model of query flows intensity depending on a time.…”

Section: Statement Of the Main Materialsmentioning

confidence: 99%

Mathematical model of server requests intensity description

Pustovoitov¹,

Kostyk²,

Kompaniiets³

et al. 2020

Bulletin of the National Technical University «KhPI» Series: New Solutions in Modern Technologies

View full text Add to dashboard Cite

The paper is devoted to the mathematical model development of non-stationary flow of requests from clients to the database in order to modulate the quality of service. The mathematical model of the queries number fluctuations to the database has the form of a regression equation and allows more accurate modeling of the connections pool size in the servlet. Connection pool is a pattern that helps to reduce responding time for queries to databases. To another hand the extra used memory wasting server resources. The task of calculating the optimal connection pool size could be solved by verity of mathematical apparatuses that demand information about intensity of stationary incoming queries flow. It is known, that real incoming queries flow is non-stationary. In the paper was suggested mathematical model of flow intensity fluctuations with daily and hourly harmonic vibrations. Statistics analyses of model adequacy was made, homogeneity of variances is checked, the significance of the coefficients of the regression equation was estimated. The obtained mathematical model describes fluctuations in the intensity of clients' requests to the servlet during the week. The mathematical model can be used to predict the load on the server or to build a simulation model of the query service system. The adequacy of the model is checked, the homogeneity of variances is checked, the significance of the coefficients of the regression equation is estimated, the adequacy of the regression equation is checked, the analysis of the autocorrelation of the residues is carried out. The results obtained in the article give further development for modeling process technologies in the field of information systems and can be used to calculate the load on the server with a non-stationary flow of requests from clients to the database.

show abstract

Section: Statement Of the Main Materialsmentioning

confidence: 99%