Trying Decision Optimization on Cloud beta: part 1, Demo
JeanFrancoisPuget 2700028FGP Visits (7888)
We just announced the open beta for our forthcoming Decision Optimization on Cloud offering. This is a major step forward in making optimization more consumable by born in the web applications. The first drop of our service enable operations research (OR) practitioners to solve their problems online via a very simple interaction. We will expose service APIs in future drops, stay tuned.
Let us look at how the service can be used right now. It assumes you have one or several optimization problems to solve, stored in files on your computer. We accept CPLEX or OPL input files.
The first thing to do is to go to Decision Optimization on Cloud page.
Scrolling down brings you to our live demo. Before registering to the service, let's try it via the demo.
After scrolling down, the screen shows three example files. We just have to select one, and drag it to the solve region to start solving it.
Solving starts when the file is uploaded:
If the solve takes too long for your taste you can abort it by clicking on the right button.
When the solve is complete, you can see the solution file, some information, and a log file,
In that case we sent a lp file which is directly solved by CPLEX. The output is a CPLEX solution file. By clicking on it you can save it to your computer.
You can then manipulate it exactly as you would do with CPLEX solution files.
Clicking on the Info tab gives your some statistics about the problem. You don't need to wait for the solve to complete to see that information.
Clicking on the log tab displays the last lines of CPLEX log.
You can then download the full log, or display the full log in a separate window by clicking the appropriate tabs at the bottom. When you are done you can discard the files on our server by clicking on the right button. It is important to do so because you are limited to 3 solves. If you want to solve a fourth problem you will need to make room by deleting one of the previous solve results.
Let us now try an OPL model. We can drag and drop one of the other two example files, for instance the one in the right. The look and feel is the same, we drag and drop the input file, which is this time an OPL project. This time the output is the same as the output you would get using OPL locally. It is a text file defined by the OPL model. The output also contains a description of the results as a tuple set, after the // SCHEMA token.
The above problem is a MIP,. The information tab and the log tab provide content, similar to our first solve.
The third problem, the one in the middle, is a constraint programming problem. We solve it the same way, by dragging the file to the solve region. We can open the info tab while the solve is running:
Once the solve is complete we can access to the result and log as before. The results are once again similar to what you would get from using OPL locally. In this case, the log file is using the CP Optimizer log format, which is different from the CPLEX log.
This concludes our first try with Decision Optimization on Cloud. The next part will deal with what you can do once you register. The first step is to get an IBM Id. If you don't have one, then start here. Once you do, then follow us on our next blog entry!