Re: Four issues why "old elephants" lack performance: Explanation sought Four issues why "old elephants" lack performance: Explanation sought

Поиск
Список
Период
Сортировка
От Andrew Sullivan
Тема Re: Four issues why "old elephants" lack performance: Explanation sought Four issues why "old elephants" lack performance: Explanation sought
Дата
Msg-id 20120227160206.GN48576@crankycanuck.ca
обсуждение исходный текст
Ответ на Re: Four issues why "old elephants" lack performance: Explanation sought Four issues why "old elephants" lack performance: Explanation sought  (Chris Travers <chris.travers@gmail.com>)
Список pgsql-general
On Mon, Feb 27, 2012 at 04:19:10AM -0800, Chris Travers wrote:
>
> Strangely, he doesn't consider PostgreSQL to be an elephant.

The last presentation I saw from Volt guys explicitly mentioned
Postgres as one of the elephants.

Also, if you read their literature carefully, you discover that the
reason there's no locking overhead is because there's no lock;
basically, if you're joining a lot, you're thrown back on
old-fashioned locks of some sort.  They also don't permit
in-transaction round trips to the application, so that source of lock
contention is also gone.

A

--
Andrew Sullivan
ajs@crankycanuck.ca

В списке pgsql-general по дате отправления:

Предыдущее
От: Adrian Klaver
Дата:
Сообщение: Re: Default PostgreSQL server encoding - Change to unicode (utf8)
Следующее
От: Tom Lane
Дата:
Сообщение: Re: explain and index scan