Re: bytea performance tweak

From: Oliver Jowett <oliver(at)opencloud(dot)com>
To: Andreas Schmitz <schmitz(at)occamlabs(dot)de>
Cc: pgsql-jdbc(at)postgresql(dot)org
Subject: Re: bytea performance tweak
Date: 2011-06-22 23:23:41
Message-ID: BANLkTimO_hQFR-5DwL5dZsMdS3XxD=WWQQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-jdbc

On 23 June 2011 02:17, Andreas Schmitz <schmitz(at)occamlabs(dot)de> wrote:

> I was wondering why the protocol uses such an encoding anyway. Wouldn't
> it be a lot faster just sending the bytes as they are?

We can request binary-format data, which for bytea is indeed just the
raw bytes, but the problem is that it's all or nothing: if we request
binary data we have to handle binary results for all datatypes, not
just bytea (because the request is per-column and generally the driver
doesn't know the column type at the point it makes the request)
Some progress has been made with that, but it's not production-ready
yet I think.

Oliver

In response to

Responses

Browse pgsql-jdbc by date

  From Date Subject
Next Message Andreas Schmitz 2011-06-23 10:06:16 Re: bytea performance tweak
Previous Message Andreas Schmitz 2011-06-22 14:22:06 Re: bytea performance tweak