Anyone know of good doc on the comparisons across the major RDMS , like the
high level performance OMGs?
Examples
- SQL Server by default is missing row level locking (OMG! of course you
can turn it on at a cost)
- Postgres , any open transaction, be it "create table foo(id int);
select * from foo; " tanks performance for everyone on the DB doing
transactions, everyone.
- Postgres - you wrap your transaction IDs and your database goes down
a day or two
- MySQL has similar issue with "create table foo(id int); select * from
foo; " where the history length list grows.
- MySQL wait events are worthless. the main wait, like 80% of all waits
is a CPU, IO and LOCK all rolled into one, yes indeed, worthless OMG
Kyle
