Re: testing HS/SR - 1 vs 2 performance

From: Simon Riggs <simon(at)2ndQuadrant(dot)com>
To: Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
Cc: Erik Rijkers <er(at)xs4all(dot)nl>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: testing HS/SR - 1 vs 2 performance
Date: 2010-04-16 21:00:09
Message-ID: 1271451609.8305.9155.camel@ebony
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, 2010-04-16 at 13:00 +0100, Simon Riggs wrote:
> Almost done

Here's the finished patch.

Internal changes only, no functional or user visible changes, so no
docs, just detailed explanatory comments.

Expectation is
* performance no longer depends upon max_connections
* recovery will be about same or slightly faster
* snapshots should be about equivalent primary/standby

Main changes are
* lock free add to sorted array (equivalent to GetNewTransactionId)
* bsearch for xid removal at commit/abort/TransactionIdIsInProgress
* faster, more compact approach to snapshots, with self-trimming
* some minor API changes to facilitate above
* new code to ignore deletion failures only when !consistent

Erik,

Could you retest please?

--
Simon Riggs www.2ndQuadrant.com

Attachment Content-Type Size
knownassigned_sortedarray.diff text/x-patch 32.0 KB

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Robert Haas 2010-04-16 21:56:04 Re: shared_buffers documentation
Previous Message Tom Lane 2010-04-16 15:10:45 Re: testing HS/SR - 1 vs 2 performance