Los Alamos Mid School (73)/Interim Report

Interim Report
http://mode.lanl.k12.nm.us/get_interim1112.php?team_id=73

Problem Definition
We study the capability of multi-agent based simulation environment. We then use the NetLogo system to design a static job scheduling simulation system.

Problem Solution
We are going to find some heuristic job scheduling algorithms that can provide a better solution for better throughout, turn-around time, less waiting time.

We have studied the NetLogo programing system and learned how to do some simple NetLogo programs.

We have started to work on my NetLogo program design. I defined agents for
 * job scheduler for various scheduling algorithms
 * client to submit random jobs
 * performance parameters such as length of a job, arriving time of a job,
 * Visual interface design

We will start to conduct some performance testings and debug my NetLogo programs on my Home PC machine. We also start to layout our simulation GUI interface.

Interim Comments
My name is Jeff Grantham and I am a graduate student in the Computer Science Department at New Mexico Tech and I'll be providing some feedback on your interim report.

The only shortcomings I can see is that the interim report is a little light on the details of what your project will be doing. It would be good to see the list of scheduling algorithms you will using in the interim report as it may have changed since you wrote the proposal and will give a good idea of the scope of your project as well as the amount of work already done and the amount to be done. Additionally, it isn't clear if the parameters for the jobs that the algorithms will be run on will be completely random or if there will some attempt to randomize the jobs in such a way that they resemble jobs you would see in the real world. Keep in mind, if you are modeling your random jobs on real world situations, you would need to do some research to determine what the real world jobs you would be modeling look like.

Otherwise, it looks like your model is a good one and that you have made very good progress on your project. The references you have included look to be a good basis for your project.

Finally, I am supposed to remind you that there will be a face to face evaluation in February.

Good luck and I look forward to seeing the completed project.

Hi Team 73, my name is Christopher Morrison. I am a challenge alumni from the years between 2003-2007. I am currently an senior at Embry Riddle Aeronautical University in Prescott, AZ studying aerospace engineering. I've just been accepted into a PhD program where I will either be involved in nuclear powered rockets or light propulsion (shining lasers at shiny discs to propel them into space). The Supercomputing Challenge really helped me and I know it will help you too. Knowing how to program and problem solve is a great trait that has helped me immensely.

I think you have a fairly straight forward task of measuring the effectiveness of various schedulers in an agent based model.

You mentioned several of the various schedulers you wanted to test. You also mentioned measuring the arrival time and length of a job. This a a good start.

What I would like to see is a test plan. I want to know

Exactly what schedulers you are testing?How you are testing them? Can you test them using some basis of unbiased comparison?How the jobs created? A cluster of jobs that are equal in length would work well with a round robin, but a round robin would not work well with a cluster of jobs with a heavily varying job length. How will you present your results to depict this?

It appears that you are making a lot of progress with what I've read from the problem solution. Keep it up and always be thinking about what your metrics (numbers you are keeping track if) mean and how to present them in a clear manner.

Good Luck