Re: BBU Cache vs. spindles

On Fri, Oct 22, 2010 at 10:28 AM, Kevin Grittner
<Kevin.Grittner@wicourts.gov> wrote:
> Rob Wultsch <wultsch@gmail.com> wrote:
>
>> has PG considered using a double write buffer similar to InnodB?
>
> That seems inferior to the full_page_writes strategy, where you only
> write a page twice the first time it is written after a checkpoint.
> We're talking about when we might be able to write *less*, not more.
>
> -Kevin
>

By "write" do you mean number of writes, or the number of bytes of the
writes? For number of writes, yes a double write buffer will lose. In
terms of number of bytes, I would think full_page_writes=off + double
write buffer should be far superior, particularly given that the WAL
is shipped over the network to slaves.

--
Rob Wultsch
wultsch@gmail.com