A review to cloud computing modeling,simulations and testing techniques

1414 Words 6 Pages
1.Intoduction
We start with the basic concepts of cloud computing and then in section 2 describes the mathematical approaches that helps in formulation of cloud models. These cloud models aims to identify cloud’s configuration settings to optimize QOS, efficiency in terms of performance and energy under available conditions.
In section 3 simulators, their architecture and features are discussed. Basically there are 2 types of simulators, i.e simulators based on software and simulators based on both software and hardware. These simulators are used in validation of those models
Section 2 (modeling) and section 3 (simulation) technology is a suitable tool for evaluating cloud performance and concerned security issues but for evaluating QOS
…show more content…
They used two penalty (overprovision and underprovision) approaches to measure imperfections in the elasticity for database system.
[3] have discussed the optimization problem by analyzing the mathematical relationship between the SLA which specifies service and number of servers with their running frequencies to optimize power usage. Basically this problem is NP-Hard.
[4] optimization problem of virtual machines allotment (developed a utility function for power saving and SLA satisfaction) and placement in actual cloud (multiple knapsack with capacity constrains ).they also discussed the management of allotment and placement using mathematical model.
[5]There are many optimizations problems such as minimizing the energy and bandwidth cost or minimizing total carbon footprint in order to govern QoS constraints. With their solution different questions can be answered like suitable location to built data center, number of servers required, routing mechanism of service requests from different servers to the data center and many more.
[6] then they discussed location diversity advantage in reducing energy cost and carbon footprint by considering workload distribution among geographically dispersed data center
[7]they have formulated another optimization problem of response time and energy consumption per job through a multi objective function. That function governs the choice between a local or remote cloud

Related Documents