Re: Hope for a new PostgreSQL era?
От | Chris Travers |
---|---|
Тема | Re: Hope for a new PostgreSQL era? |
Дата | |
Msg-id | CAKt_ZfszcGeiHyG1UapgyMuOFA=efYzu8tEw5tUt5Rwmo0gicA@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Hope for a new PostgreSQL era? (Merlin Moncure <mmoncure@gmail.com>) |
Список | pgsql-general |
On Thu, Dec 8, 2011 at 6:54 AM, Merlin Moncure <mmoncure@gmail.com> wrote: > There is simply no comparing mysql's backend programming features with > those of postgres. Postgres is a development platform in a box, mysql > is not. > This and there are annoyances I have run into with MySQL, such as deadlocks.... on a system with only one running transaction.... I traced this problem eventually to multi-row inserts and concluded it was a race condition in threaded processing. As I say, any db can and should allow conflicting transactions to deadlock. It takes special talent to allow transactions to deadlock against *themselves*...... Interestingly I have customers who run both MySQL and PostgreSQL and have had similar issues with transactions deadlocking against themselves on MySQL, so I know I am not unique here. MySQL, I am sorry to say, is not catching up to PostgreSQL at least from where I stand except for the few very specific workloads that its query cache are designed to work with. Best Wishes, Chris Travers
В списке pgsql-general по дате отправления: