Optispot: Minimizing Application Deployment Cost Using Spot Cloud Resources

1. Attached files: This archive contains 1800 MATLAB files, each one containing the results of a single experiment.<br> The name of each file follows the following format: A_B_C_D_E_F_G.mat Where the fields A, B, C, D, E, F, and G are described as follows. A: number of users.<br> Considered values are: 1000, 2000, 5000, 10000. B: maximum response time in milliseconds.<br> Considered values are: 60, 80, 100, 200. C: overbid time cap in hours.<br> Considered values are: 5, 20, 80, 0 (note: 0 is a code used to express infinite hours). D: Amazon region.<br> Considered values are: us-east, eu-west. E: Operating system.<br> Considerede values are: Windows, Linux. F: Optimization algorithm.<br> Considered values are: heuristic (which is OptiSpot), fmincon. G: Experiment seed.<br> Considered values are from 1 to 30 2. Data format: MATLAB data format, can be loaded from MATLAB using the following command: results = load(filename); results is defined as a structure with the following fields: results.cost<br> Type: scalar, positive real number.<br> Desc: hourly cost in US dollars. <br> results.time<br> Type: scalar, positive real number.<br> Desc: total time (in seconds) needed by the algorithm to compute the solution. results.evaluations<br> Type: scalar, positive integer number.<br> Desc: number of constraints evaluations needed by the algorithm to compute the <br> solution. <br> results.d<br> Type: matrix, non negative positive real number. <br> Desc: association matrix between rented resources (columns) and application <br> components (rows). The sum of all the elements of this matrix is equal to<br> the ECUs used by the application.