In the old days, the price for IT services was formed in a pretty standardized way. Network services had an explicit usage price per Kbit/ sec. The range of provided IT services have been growing very fast and have reached new dimensions of complexity. From infrastructure pricing to web-enabled application availability and performance nowadays the old rules for defining service pricing is not applicable any more. Today it is difficult or sometime even impossible to associate the provided service levels with the cost related to the processes of operation, maintenance and the capital cost behind it. The old measures of dollars per Kbit/sec cannot be the right measure any more. For example setting the price for one transaction per second does not define at all what is needed to perform two transactions per second. What should be a penalty in the case of breaking SLAs? Where is the golden mean here? How can it be found?
In the paper we consider solving such kind of tasks by means of simulation. The simulation model can serve as a decision support tool for predicting (calculating) the price for a particular service, which is being provided. Using such models by service providers will ensure them with high probability of cost recovery and profit, competitiveness of prices, and encouragement of client behaviors that will enhance the services’ efficiency. The implementation and usage costs of the models are low as soon as it can be done in easy way and without special knowledge in simulation.
IT Infrastructure parameters optimization