Day 9 results: B-tree vs LSM, the write-order gap
Measured on: (machine, OS, Python version)
Output
Paste part 1, part 2, part 3 and the scoreboard here.
One sentence per prediction
- P1 (B-tree random slowdown):
- P2 (LSM write ratio random / sequential):
- P3 (LSM sorted runs):
- P4 (LSM read places on a miss):
The B-tree’s weakness, in my own words
Same rows, same keys, only the order changed. How much slower was random, and what was the tree actually doing:
The LSM’s bargain
Why was the append write path flat for sequential vs random, and what did it cost on the read side:
The number that matters
Random B-tree inserts were __ slower than sequential. Append-only writes came out __ (flat). Why that gap sends write-heavy systems to an LSM: