Sas System

This project is designed as an independent study CS 999 course in SAS programming arranged for the fall 2008 semester at UW-M.  I am taking the SAS base certification exam in July.  One goal is for me to gain experience writing SAS programs.  My proposal is a one credit course in which I will use Microsoft Management Console 2.0 (Perfmon) to generate flat files containing system metrics, including memory, disk space, CPU, and .Net counter metrics.  I will install SAS on my system.  I will then use SAS to build data sets in order to analyze the performance data, practicing using SAS SQL and SAS functions such as proc means and proc freq.  I also will experiment with graphing data, both by using proc chart and by generating XML that I will display on the web using XML/SWF Charts Flash technology.

 

I will pull data files using Perfmon collecting data at 5 second intervals for 30 minutes.  I will pull four such files, one with the system idle, one with McAfee virus scan software running, one with Dragon voice recognition software running, and one with Netflix software playing a movie.

 

In order to download SAS on my system I will go to the following URL: http://www.sas.com/govedu/edu/programs/oda_account.html.  I will then enter the course information and be able to obtain a $60 educational license.

 

The first task in terms SAS programming is to read the data into a data set.  I will write a program to perform this task.

 

Another task is to use the SAS proc means function in order to generate summary statistics for disk space, CPU utilization, and memory utilization. The summary statistics include minimum and maximum values, standard deviation, and mean values.  I will write a program to perform this task.

 

And based on the results of the proc means summary statistics we could then generate a frequency distribution for those specific processes that use high amounts of system resources.  I will write a program to perform this task.

 

One useful way to analyze the data is to pull out those processes that use more than 90% of key system resources, such as CPU, disk space, or memory.  I will write a program to perform this task.

 

Another thing we could do would be to analyze what level of sampling frequency would be sufficient to avoid missing spikes in the data.  I will develop SAS code to run a T test to determine whether a 1 minute sampling interval would be sufficient.

 

In order to visualize the data  more clearly it is helpful to graph it.  In order to generate graphs of our data the following SAS chart commands will be used, as well as the SAS capability to produce an XML file which can then be displayed graphically on the web using Flash technology.  (Adobe Flash, previously called Shockwave Flash and Macromedia Flash, is a set of multimedia technologies developed and distributed by Adobe Systems and earlier by Macromedia.)

 

In order to generate readable graphs it will not be possible to graph every data point since our data was pulled at five second intervals, this would create graphs that were too messy. Code will be developed to pull out data points at one minute intervals to create more readable graphs.

 

Comparison to other statistical packages, such as JMP, Statistica, and Excel will be made, as well as to another graphing tool, such as Chart FX.

 

The grade will be based on the quality of a final paper.  Progress will be checked bi-weekly throughout the semester.

 

Timeline outlining the progress of the work. 

 

Here is a monthly breakdown of what will be done:

 

By the end of September have the basic data collection and analysis SAS code for creating the data sets, generating the summary statistics, generating a frequency distribution, doing a T test, generating charts and generating web-based XML/SWF Charts graphs completed.

By the end of October develop additional SAS code to reinforce the Sharpening your SAS Skills book chapters on accessing data, creating data structures, and managing data.  These chapter outlines are reproduced below.

By the end of November develop additional SAS code to reinforce the Sharpening your SAS Skills book chapters on generating reports and handling errors.

By the completion of the course at the end of the semester in December complete comparisons between SAS and Excel, JMP, Statistica, and Chart FX.  Complete final paper.

 

            The textbook for the course will be Sharpening your SAS Skills by S. Gupta.  SAS programs will be developed to practice using each of the key concepts mentioned in the table of contents for this book reproduced below.

 

 

 

2 SAS System Controller