Programming Assignment #1

A DePaul Computer Science student is interested in comparing the TPC-D benchmark score for two popular relational database management systems. The products are "Oracle8" and "SQL Server 6.5".

She is particularly interested in performance for machines running Microsoft "Windows NT Server". She wishes to challenge a colleague who argues that, for TPC-D, there is no difference in performance between these products. She believes that "SQL Server 6.5" is optimized to run under NT.

She decides to collect benchmark scores for several machines. She identifies the necessary machines, installs NT, as well as Oracle8 and SQL Server, and then executes the TPC-D benchmark on each machine. She has asked you to analyze her benchmark scores.

Each observation consists of four values:

• CPU type
• CPU Architecture (RISC, CISC)
• Oracle TPC-D Score (throughput metric)
• SQL Server TPC-D Score (throughput metric)

Conduct a thorough analysis of these data. Ensure that you do the following:

1. Analysis:
   • Justify the procedure you use to analyze these data.
   • State the appropriate NULL and alternative hypotheses.
   • Determine if these data provide sufficient evidence to indicate that SQL Server is superior to Oracle8 for this type of workload.

     Note: Use the format at the end of "Knafl - chapter 3" as a model for your analysis.

   
   
2. Code:
   • Use meaningful variable names and dataset names.
   • Use label statements where necessary to give your output sensible headings.
   • Include appropriate title statements.
   • Include "proc print" in your program to print input values.
   • Use "proc rank" and "proc plot" to produce a normal plot.

Note: Turn in your SAS code, SAS output, SAS log file and your analysis.