基于排队petri网的oltp系统性能建模方法分析与实现-analysis and implementation of oltp system performance modeling method based on queuing petri net.docx

AbstractWith the continuous development of information technology, the OLTP service system is widely used in commercial banks, telecommunications companies, stock exchanges, airlines and other large commercial financial institutions and enterprises. Once system crash or lack of capacity, it will affect the normal running of the financial industry, as to cause significant economic losses to the state. Whether the OLTP application service system is able to provide a stable and efficient services to meet the needs of enterprise applications in realistic environments become initial problem that the industry need urgently resolve.To predict and evaluate the OLTP performance of OLTP service system, in order to proactively identify bottlenecks that exists in the system is particularly important in the increasingly complex application environments. Queueing theory and stochastic Petri Netsis used for modeling and simulatingof the service system to establish the model of a mathematical equation. By simulating the model, it will obtain performance indicators that reflecting the target system in the given input parameters. Queueing Petri Nets inherit and develop the advantages of a queueing network models and stochastic Petri Nets, can either qualitatively describe the interactions between the system multi-layer components or quantitatively analyze the service performance.The article established performance model for a typical OLTP applications using queuing Petri Net, model made the choice of banking intermediary business as a typical application load, the model describednot only the the service law and service capabilities of load equipment, load balancing equipment, database servers, disk array and other components in OLTP application systems, but also the interaction behavior and interaction rules of various components in detail. Finally, the graphical modeling tool QPME is used in establishing and simulating of the model, to obtain the throughput capacity,service respons