Re: Thousands of schemas and ANALYZE goes out of memory - Mailing list pgsql-general

From Chris Angelico
Subject Re: Thousands of schemas and ANALYZE goes out of memory
Date
Msg-id CAPTjJmoi2wR9YfTp+9igv8c=6RvGugLpub9wHh7gWq6n7SWPhw@mail.gmail.com
Whole thread Raw
In response to Re: Thousands of schemas and ANALYZE goes out of memory  (Jeff Janes <jeff.janes@gmail.com>)
Responses Re: Thousands of schemas and ANALYZE goes out of memory  ("Hugo <Nabble>" <hugo.tech@gmail.com>)
List pgsql-general
On Wed, Oct 3, 2012 at 10:09 AM, Jeff Janes <jeff.janes@gmail.com> wrote:
> On Tue, Oct 2, 2012 at 10:38 AM, Hugo <Nabble> <hugo.tech@gmail.com> wrote:
>>> That might be the problem.  I think with 32 bits, you only 2GB of
>>> address space available to any given process, and you just allowed
>>> shared_buffers to grab all of it.
>>
>> The address space for 32 bits is 4Gb.
>
> I had thought the highest bit was not usable, but maybe that was just
> a Windows thing.

Losing the highest bit is common when something treats a number as
signed, but the <4GB memory issue isn't that. It's because other areas
of memory need to be addressable (devices claim certain areas), and
all that has to be factored into the total 4GB addressable space.
Usually you get >3GB, somewhere around the 3.5GB mark.

ChrisA


pgsql-general by date:

Previous
From: Jeff Janes
Date:
Subject: Re: Thousands of schemas and ANALYZE goes out of memory
Next
From: Royce Ausburn
Date:
Subject: invalid page header in block...