I am excited to announce the availability of the GA release of our new product Percona XtraDB Cluster.

Percona XtraDB Cluster is a High Availability and Scalability solution for MySQL Users and is based on Percona Server 5.5.20.

With this release we make clustering very easy and affordable for everyone. You can convert your existing MySQL replication setup into a cluster in just a couple minutes.

Tomorrow, 4-April-2012, I will be giving a webinar where I will explain what Percona XtraDB Cluster is and how does it work, you still have time to register today.

I expect XtraDB Cluster and Galera will be a big topic on coming MySQL User Conference, both Codership and Percona development team will be there, you will have a chance to ask us any questions.

Related talks:

Percona XtraDB Cluster provides:

  • Synchronous replication. Transaction either commited on all nodes or none.
  • Multi-master replication. You can write to any node.
  • Parallel applying events on slave. Real “parallel replication”.
  • Automatic node provisioning.
  • Data consistency. No more unsyncronised slaves.

Percona XtraDB Cluster is fully compatible with MySQL or Percona Server in the following meaning:

  • Data compatibility. Percona XtraDB Cluster works with databases created in MySQL / Percona Server
  • Application compatibility. There is no or minimal application changes required to start work with Percona XtraDB Cluster

Percona XtraDB Cluster is based on Percona Server with XtraDB and includes Write Set REPlication patches. It uses the Galera library, version 2.x, a generic Synchronous Multi-Master replication plugin for transactional applications.

The Galera library is developed by Codership Oy company and
Galera 2.x supports such new features as:

  • Incremental State Transfer, especially useful for WAN deployments
  • RSU, Rolling Schema Update. Schema change does not block operations against table

This is an General Availability release. We did our best to eliminate bugs and problems during alpha and beta testing release, but this is a software, so bugs are expected. If you encounter them, please report to our bug tracking system.

Links:

Previous posts on this topic:

Happy clustering!

12 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Dave Juntgen

Vadim – wonderful news and great work.

I’m curious to how you came to a GA so quickly after releasing the Alpha and then Beta in a 12 month period. What type of QA and workloads have you tested?

Regards,

–Dave

0
Brian

Per the release notes, there are some LOAD DATA related limitations. It says “huge” LOAD DATA will be rejected. What is considered “huge”? We do quite a lot with LOAD DATA in our system, and sometimes the files can be a couple GB in size.

0
Vadim Tkachenko

Dave,

That is a valid question, let me answer on it from different points.

XtraDB Cluster is integration of Percona Server and Galera library. Both Percona Server and Galera are in GA state,
so the main focus of testing was on proper integration and packaging.
Galera was in in my focus since 2009, so this is really 3 years of testing and evaluations for me. At this point I believe it has good quality.

We used following tools for evaluations: randgen (lp:randgen), kewpie (lp:kewpie) and for stressing and performance: sysbench and tpcc-mysql.

0
Art van Scheppingen

That’s great news!
Why so shortly before the conference? Now I don’t have enough time to play with it before attending your sessions. 🙁

0
Vadim Tkachenko

Brian,

I was told that LOAD DATA works differently now.
Quote from Seppo: “LOAD DATA INFILE processing will commit every 10K rows. So large transactions due to LOAD DATA will be split to series of small transactions”.

We should fix documentation.

0
Manjari Shukla

Hi Vadim,
G8 news……. thnx for shareing..

0
Luke Ludwig

Vadim,

Nice work! I’m excited about the future of Percona Xtradb Cluster as it becomes more mature. I’m curious about Percona Xtradb Cluster from the perspective of running across multiple AWS regions, for example having 2 masters in the US Eastern region and 1 in the US Western region. As I understand it, due to the synchronous replication all writes would be affected by the network latency between these regions. Is Percona Xtradb Cluster not designed or not ideal for this type of setup?

Luke

0
Gene Vayngrib

Vadim, I second Luke’s request. It would be extremely helpful for many of us to get your recommendations on how to best deploy Xtradb Cluster in AWS environment. It would also be great to get some idea on what throughput increase we can expect considering EC2 and EBS constraints. Our company is working on adding NoSQL (AWS DynamoDB) and AWS EMR components to the setup which is now exclusively based on Percona’s MySQL server. The more we can extract from MySQL without relegating to the extremely limiting NoSQL setup, the simpler our applications will be.

Thank you,
Gene

0
karim soft

Hi,

I use percona server 5.5.20, I would be testing xtradb cluster with three nodes. I went through your documentation provided with the release, and noticed the configuration, but i did not understand where did all of the configuration go for example default_storage_engine, query_cache_size, e.t.c…. Is it something built in or we just don’t need to define it for the cluster. I am new in this field and being a startup company cannot afford to get paid support from percona, but at-least you can answer this one or provide a link to sample configuration. Your tool do not provide config for xtradb cluster.

thanks..

karim…..

0
Vadim Tkachenko

I deleted comment from someone with name “mr unhappy”.
The comment did not provide constructive feedback.
If you have problem with software – please report bugs to
https://bugs.launchpad.net/percona-xtradb-cluster
and we are doing our best to answer on questions about XtraDB Cluster in our mailgroup
http://groups.google.com/group/percona-discussion

0
Doug Bell

Can you clarify DDL handling issues? The documentation only says “DDL statements are problematic and may stall cluster. Later, the support of DDL will be improved, but will always require special treatment.” Our current implementation has 1,500 databases with 53,000 tables, so table add, drop, and modify actions happen rather frequently.

0
Benjamin Mantilla

Vadim – wonderful news and great work.

0