OptiSpot: Minimizing Application Deployment Cost using Spot Cloud Resources

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