Horton

Re: Clock sweep not caching enough B-Tree leaf pages?

Peter Geoghegan <pg@heroku.com>

View thread View on postgr.es
From: Peter Geoghegan <pg@heroku.com>
To: Pg Hackers <pgsql-hackers@postgresql.org>
Cc: Jim Nasby <jim@nasby.net>, Robert Haas <robertmhaas@gmail.com>, Andres Freund <andres@2ndquadrant.com>, Merlin Moncure <mmoncure@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>
Date: 2014-04-25T17:45:53Z
Lists: pgsql-hackers 
I've now done a non-limited comparative benchmark of master against
the patch (once again, with usage_count starting at 6, and
BM_MAX_USAGE_COUNT at 30) with a Gaussian distribution. Once again,
the distribution threshold used was consistently 5.0, causing the
patched pgbench to report for each test:

transaction type: Custom query
scaling factor: 5000
standard deviation threshold: 5.00000
access probability of top 20%, 10% and 5% records: 0.68269 0.38293 0.19741

Results are available from:

http://postgres-benchmarks.s3-website-us-east-1.amazonaws.com/3-sec-delay-gauss/

-- 
Peter Geoghegan

Commits

  1. Replace the BufMgrLock with separate locks on the lookup hashtable and