I talk with lot of people who are really interested in Percona XtraDB Cluster (PXC) and mostly they are interested in PXC as a high-availability solution. But, what they tend not to think too much about is if moving from async to synchronous replication is right for their application or not. Facts about Galera replication [...]
Benchmarking Percona Server TokuDB vs InnoDB
May 7, 2013 By 29 Comments
After compiling Percona Server with TokuDB, of course I wanted to compare InnoDB performance vs TokuDB. I have a particular workload I’m interested in testing – it is an insert-intensive workload (which is TokuDB’s strong suit) with some roll-up aggregation, which should produce updates in-place (I will use INSERT .. ON DUPLICATE KEY UPDATE statements [...]
Follow these basics when migrating to Percona XtraDB Cluster for MySQL
May 1, 2013 By 9 Comments
Galera/Percona XtraDB Cluster (PXC) for MySQL is a hot thing right now and some users jump right in without enough testing. Consequently, they’re more likely to either suffer failure or issues that prevent them from moving forward. If you are thinking of migrating your workload to Percona XtraDB Cluster, make sure to go through these [...]
More on MySQL transaction descriptors optimization
April 26, 2013 By Leave a Comment
Since my first post on MySQL transaction descriptors optimization introduced in Percona Server 5.5.30-30.2 and a followup by Dimitri Kravchuk, we have received a large number of questions on why the benchmark results in both posts look rather different. We were curious as well, so we tried to answer that question by retrying benchmarks on [...]
How to recover table structure from InnoDB dictionary
April 22, 2013 By 4 Comments
To recover a dropped or corrupt table with Percona Data Recovery Tool for InnoDB you need two things: media with records(ibdata1, *.ibd, disk image, etc.) and a table structure. Indeed, there is no information about the table structure in an InnoDB page. Normally we either recover the structure from .frm files or take it from [...]
Percona XtraDB Cluster 5.5.30-23.7.4 for MySQL now available
April 17, 2013 By 2 Comments
Percona is glad to announce the release of Percona XtraDB Cluster 5.5.30-23.7.4 for MySQL on April 17, 2013. Binaries are available from the downloads area or from our software repositories. New Features: Percona XtraDB Cluster has implemented initial implementation of weighted quorum. Weight for node can be assigned via pc.weight option in the wsrep_provider_options variable. [...]
How to create/restore a slave using GTID replication in MySQL 5.6
February 8, 2013 By 7 Comments
MySQL 5.6 is GA! Now we have new things to play with and in my personal opinion the most interesting one is the new Global Transaction ID (GTID) support in replication. This post is not an explanation of what is GTID and how it works internally because there are many documents about that: http://dev.mysql.com/doc/refman/5.6/en/replication-gtids-concepts.html One [...]
Implications of Metadata Locking Changes in MySQL 5.5
February 1, 2013 By 9 Comments
While most of the talk recently has mostly been around the new changes in MySQL 5.6 (and that is understandable), I have had lately some very interesting cases to deal with, with respect to the Metadata Locking related changes that were introduced in MySQL 5.5.3. It appears that the implications of Metadata Locking have not [...]
Knowing what pt-online-schema-change will do
November 1, 2012 By 1 Comment
pt-online-schema-change is simple to use, but internally it is complex. Baron’s webinar about pt-online-schema-change hinted at several of the tool’s complexities. Consequently, users often want to know before making changes what pt-online-schema-change will do when it runs. The tool has two options to help answer this question: –dry-run and –print. When ran with –dry-run and –print, pt-online-schema-change changes nothing [...]
Edge-case behavior of INSERT…ODKU
October 27, 2012 By 6 Comments
A few weeks back, I was working on a customer issue wherein they were observing database performance that dropped through the floor (to the point of an outage) roughly every 4 weeks or so. Nothing special about the environment, the hardware, or the queries; really, the majority of the database was a single table with [...]
