News for hackers. Stuff that matters.
Re:Batch size one billion: SQLite insert speedups, from the ... by klaussilveira on Tuesday September 26, 2023 @11:01AM (#37660399)
That's a fantastic analysis. I wonder how PostgreSQL performs?
Re:Batch size one billion: SQLite insert speedups, from the ... by eatonphil on Tuesday September 26, 2023 @11:05AM (#37660459)
I've done some comparisons for inserts of various data sizes on mariadb, postgres, and sqlite without indexes.

I didn't have the patience of OP though to push it to 1B.

https://github.com/eatonphil/databases-intuition

Re:Batch size one billion: SQLite insert speedups, from the ... by riku_iki on Tuesday September 26, 2023 @04:23PM (#37665320)
looks like PGSQL 2x slower for uploading.
Re:Batch size one billion: SQLite insert speedups, from the ... by eatonphil on Tuesday September 26, 2023 @04:26PM (#37665362)
Looks like I'm not alone. But I don't, and don't think Mark does (he's a MySQL guy I believe), claim to be a Postgres expert.

http://smalldatum.blogspot.com/2023/09/trying-out-orioledb.h...

Re:Batch size one billion: SQLite insert speedups, from the ... by voidstarcpp on Tuesday September 26, 2023 @12:35PM (#37661841)
PostgreSQL supports heap tables, which should blow away SQLite's mandatory clustered index tables in an unindexed insert test.
Re:Batch size one billion: SQLite insert speedups, from the ... by chmaynard on Tuesday September 26, 2023 @01:39PM (#37662933)
In SQLite, would declaring the table without a rowid obviate the index? For example:

    CREATE TABLE IF NOT EXISTS integers
    (
       value INTEGER NOT NULL, 
    )
    WITHOUT ROWID;