Re: performance question: protocol v2 vs v3

From: Albe Laurenz <laurenz(dot)albe(at)wien(dot)gv(dot)at>
To: "Craig Ringer *EXTERN*" <craig(at)2ndquadrant(dot)com>, "Guillaume Cottenceau *EXTERN*" <gc(at)mnc(dot)ch>, "pgsql-jdbc(at)postgresql(dot)org" <pgsql-jdbc(at)postgresql(dot)org>
Subject: Re: performance question: protocol v2 vs v3
Date: 2014-12-03 11:41:39
Message-ID: A737B7A37273E048B164557ADEF4A58B17DA76B4@ntex2010a.host.magwien.gv.at
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-jdbc

Craig Ringer wrote:
> On 11/25/2014 06:45 PM, Albe Laurenz wrote:
> > An explanation is maybe that prepared statements are handled
> > using the extended query protocol, which only exists in v3.
> > With the extended query protocol, each statement is first parsed,
> > then parameters are bound to the statement and then the statement
> > is executed, leading to three client-server round trips.

> While Parse/Bind/Execute are separate steps, it is not correct to say
> that this means three round trips.
>
> Take a look at the conversation on the wire in Wireshark. You'll see
> that generally, the Parse, Bind, Describe, Execute and Sync messages are
> all actually sent in the same packet.

I didn't know that, thanks for the clarification.

Yours,
Laurenz Albe

In response to

Responses

Browse pgsql-jdbc by date

  From Date Subject
Next Message Craig Ringer 2014-12-03 13:44:21 Re: performance question: protocol v2 vs v3
Previous Message Craig Ringer 2014-12-03 11:33:40 Re: Having issues with tests