Blocking factor of Data Blocks = No of records per block
= $\left \lfloor \frac{Data block size}{record size} \right \rfloor$
= $\left \lfloor \frac{1024}{100} \right \rfloor$ = 10
Total no of Data Blocks = Total no of Records / Blocking factor
= $\left \lceil \frac{10000}{10} \right \rceil$ = 1000
Amy’s implementation using Sparse Index :-
Index entry size = Search key size + Pointer to the Data block in which the particular record is located
=> 9 + 7 = 16 Bytes
Considering Index Block size = Data Block size;
Blocking factor of Index Block = No of index entries per block
= $\left \lfloor \frac{Data block size}{index entry size} \right \rfloor$
= $\left \lfloor \frac{1024}{16} \right \rfloor$ = 64
Sparse index ensures an index entry in index block for each Data block
Total no of Index entries = Total no of Data Blocks
No of Index blocks required using Sparse implementation = Total no of index entries / Blocking factor of Index block
= $\left \lceil \frac{1000}{64} \right \rceil$ = 16
Bob’s implementation using Dense Index :-
Dense index ensures an index entry in index block for each Record of all Data Blocks
Total no of Index entries = Total no of Records
No of Index blocks required using Dense implementation = Total no of index entries / Blocking factor of Index block
= $\left \lceil \frac{10000}{64} \right \rceil$ = 157
No of blocks saved by Amy considering only 1st level index = 157 – 16
= 141
