Open jonyejin opened 11 months ago
Nodes use page ids to reference other nodes in the index. The DBMS must get the memory location from the page table during traversal. If a page is pinned in the buffer pool, then we can store raw pointers instead of page ids. This avoids address lookups from the page table.
๐ ๊ฐ์
์๊ฐ
: 2023-08-11(๊ธ) 22:00์๋ฃ
: 8๊ฐ B+Tree Indexes์ฐธ์ฌ์
: ํญ๊ท, ๊ทธ๋ฃจํธ, ๋ ์ฌ, ๋๊ฐโ๏ธ ์ค๋น
์๋ ์์์ ๋ณต์ฌํด ์คํฐ๋ ์ค๋น๋ฅผ ์๋ฃํด์ฃผ์ธ์!