2. Consider a relation P with attributes ABCDEFGH and functional dependencies T. T= { ABC, CABD,...

 

Transcribed Image Text:

2. Consider a relation P with attributes ABCDEFGH and functional dependencies T. T= { ABC, CABD, CFDE, EB, BFEC, BDA} (a) Compute a minimal basis for T. In your final answer, put the FDs into alphabetical order. Within a single FD, this means stating an FD as XY A, not as YX A. Also, list the FDs in alphabetical order ascending according to the left-hand side, then by the right-hand side. This means, WX A comes before WXZ A which comes before WXZ B. (b) Using your minimal basis from the last subquestion, compute all keys for P. (c) Employ the 3NF synthesis algorithm to obtain a lossless and dependency-preserving decomposition of relation P into a collection of relations that are in 3NF. Do not "over normalize". This means that you should combine all FDs with the same left-hand side to create a single relation. If your schema includes one relation that is a subset of another, remove the smaller one. (d) Does your schema allow redundancy? Explain how you know that it does or does not. 2. Consider a relation P with attributes ABCDEFGH and functional dependencies T. T= { ABC, CABD, CFDE, EB, BFEC, BDA} (a) Compute a minimal basis for T. In your final answer, put the FDs into alphabetical order. Within a single FD, this means stating an FD as XY A, not as YX A. Also, list the FDs in alphabetical order ascending according to the left-hand side, then by the right-hand side. This means, WX A comes before WXZ A which comes before WXZ B. (b) Using your minimal basis from the last subquestion, compute all keys for P. (c) Employ the 3NF synthesis algorithm to obtain a lossless and dependency-preserving decomposition of relation P into a collection of relations that are in 3NF. Do not "over normalize". This means that you should combine all FDs with the same left-hand side to create a single relation. If your schema includes one relation that is a subset of another, remove the smaller one. (d) Does your schema allow redundancy? Explain how you know that it does or does not.