I wanted to understand how row-store key prefix compression works for integer data types in WiredTiger. As part of the source code, there is a packing done for each data type and specifically for integer data types, a variable-sized packing is done based on signed/unsigned integers and the range of the integer value. I wanted to understand this and how it is used in prefix compression.
Thank you,
Shreya
I want to understand how packing enables lexicographic comparison for integers.It would be great if there are any reference documents for this.