Tuesday, October 24

How to install multiple PostgreSQL servers on RedHat Linux

Gabriele's PlanetPostgreSQL, PostgreSQL
If you have a Linux server of the RedHat family (inclusing CentOS and Fedora), you might envy the way Debian/Ubuntu distributions handle PostgreSQL clusters management. Although it is not easy to install different PostgreSQL versions on the same RedHat Linux server using RPMs, it is much simpler to install several instances of PostgreSQL (servers) and, at the same time, take advantage of the services infrastructure. (more…)

The Return of XFS on Linux

Greg's PlanetPostgreSQL, PostgreSQL
If you're running Linux, and particularly if you're running a database on Linux, it's been hard to recommend any filesystem other than plain old ext3 in recent years.  Some of the alternatives that looked interesting at one point--jfs, ReiserFS--are completely abandoned at this point.  The one that has been almost viable for some time now is XFS, originally an SGI projecs.  And it's back to being in the limelight again this week.XFS had suffered from a number of problems in the past.  Since it was designed for stable hardware, it wasn't as robust on standard cheap PC hardware at first; quite a bit of that was just cleaned up two years ago.  It had this odd problem with zeroed files that scared some people off.  It was treated as a second-class citizen in (more…)

AMD, Intel, and PostgreSQL

Greg's PlanetPostgreSQL, PostgreSQL
A few weeks ago I presented an updated 2010 version of my talk on database hardware benchmarking at PG East; slides available from our talks page.  CPU and memory performance are particularly important for a PostgreSQL database, because every individual query runs as a single process.  Therefore, the speed of your fastest core determines how fast any one query can execute at, and in modern systems that's quite likely to bottleneck based on memory speed.One of the things that's obvious from recent memory speed results is that all of AMD's processors have been stuck in a distant second place for almost 18 months now.  While AMD continues to use DDR2-800, Intel's "Nehalem" processors, shipping in volume since early 2009, have been adopting increasingly fast DDR3 in good (more…)

Installing Greenplum Single Node Edition on Amazon’s EC2

Gabriele's PlanetPostgreSQL, Greenplum
I have been thinking for a while now about adding Greenplum support to an open-source application for web analytics that I wrote a few years ago, which is called ht://Miner and uses PostgreSQL. In order to do this, I need a multi-CPU environment. While still waiting to get our new servers installed here in our data centre in Italy, I decided to look at Amazon's Elastic Compute Cloud (EC2) infrastructure. My intention is to do some benchmarking and spot the main differences in terms of performances between Greenplum Single Node Edition and PostgreSQL 8.4, my favourite DBMS. If you wish to follow this article, you need to have an Amazon AWS account with a valid credit card. Do not worry, this test will only cost you a couple of dollars! (more…)

PGEast, Hardware Benchmarking, and the PG Performance Farm

Greg's PlanetPostgreSQL, PostgreSQL, United States News
Today is the deadline for the special room rate at the hotel hosting this month's PostgreSQL Conference East 2010.  If you've been procrastinating booking a spot at the conference, as of tomorrow that will start costing you.My talk is on Database Hardware Benchmarking and is scheduled for late afternoon on the first day, Thursday March 25th.  Those who might have seen this talk before, either live at PGCon 2009 or via the video link available there, might be wondering if I'm going to drag out the same slides and talk again.  Not the case; while the general philosophy of the talk ("trust no one, run your own benchmarks") stays the same, the examples and test mix suggested have been updated to reflect another year worth of hardware advances, PostgreSQL work, and my own (more…)

Trade-offs in Hot Standby Deployments

Greg's PlanetPostgreSQL, PostgreSQL
The new Hot Standby feature in the upcoming PostgreSQL 9.0 allows running queries against standby nodes that previously did nothing but execute a recovery process. Two common expectations I've heard from users anticipating this feature is that it will allow either distributing short queries across both nodes, or allow running long reports against the standby without using resources on the master. These are both possible to do right now, but unless you understand the trade-offs involved in how Hot Standby works there can be some unanticipated behavior here. Standard Long-running Queries One of the traditional problems in a database using MVCC, like PostgreSQL, is that a long-running query has to keep open a resource--referred to as a snapshot in the current Postgres implementation--to (more…)