From 57c3f359c24d8cb3de1d36a6612ad65763173519 Mon Sep 17 00:00:00 2001 From: Alexander Kiel Date: Mon, 20 Feb 2023 20:15:56 +0100 Subject: [PATCH] foo --- docs/implementation/database.md | 1 + 1 file changed, 1 insertion(+) diff --git a/docs/implementation/database.md b/docs/implementation/database.md index 73161c913..37e411fca 100644 --- a/docs/implementation/database.md +++ b/docs/implementation/database.md @@ -14,6 +14,7 @@ The idea behind an immutable database is, that the whole database content at a p +-------------+ +-------------+ ``` + Database values are not copied entirely from one version to the next. Instead, like in persistent data structures, structural sharing is used. As such each database value can be seen as a complete copy of the database from the outside, but at the inside, the implementation is efficient enough to be feasible nowadays. **Note:** In contrast, relational databases, which where designed in the 80's, use an update in-place model, because storage was expensive then.