Every few months, I get the fun job of announcing what’s new in TokuDB®, but this time is special. With Version 7, TokuDB for MySQL and MariaDB is going open source.
The free Community Edition is fully functional and fully performant. It has all the compression you’ve come to expect from TokuDB. It has hot schema changes: no-down-time column insertion, deletion, renaming, etc., as well as index creation. It has clustering secondary keys. We are also announcing an Enterprise Edition (coming soon) with additional benefits, such as a support package and advanced backup and recovery tools.
Making TokuDB open source is a natural next step for Tokutek’s involvement in the MySQL community. So far, Tokutek has been involved in the community in many ways:
- We’ve contributed a number of MySQL and MariaDB patches, as well as source code for our parallel bulk loader.
- Our iiBench benchmark has been enhanced greatly over the years through community contributions.
- Bradley C. Kuszmaul, a co-founder of Tokutek and chief architect, has been a member of the IOUG council since it was formed in 2011.
- Gerry Narvaja, who leads technical services at Tokutek, is the co-host of the popular MySQL Community podcast OurSQL and won the MySQL Community Contributor of the Year 2012 award at the recent Percona MySQL Users Conference.
- Members of Tokutek’s technical team have been frequent speakers on general industry topics such as Indexing, Scaling, and Benchmarking at multiple MySQL events, including Percona Live, MySQL Connect, and various meetups.
TokuDB v7 maintains all our established advantages: fast trickle load, fast bulk load, fast range queries through clustering indexes, no fragmentation, and full MySQL/MariaDB compatibility for ease of installation.
In addition, there are plenty of other performance improvements included in this version. For starters, TokuDB v7 adds support for Direct I/O. Also, you asked for it, you got it: TokuDB v7 has significantly enhanced Engine Status information.
For details on updates to pricing and supported MySQL and MariaDB versions, please see our FAQ.
To learn more about TokuDB:
- Download executables here.
- The source code is available on GitHub.
- Attend a Webinar overview of TokuDB v7 on May 2nd.
- Our public IRC channel is #tokutek on freenode.net. We have google groups at tokudb-dev and tokudb-user.
- Read our blog.
- Learn more about Fractal Tree Indexing.
- Join our TokuDB mailing list to be notified about future product updates and related events.
- Stop by Tokutek booth # 114 or hear one of our presentations at the Percona Live and SkySQL Conferences on April 22-26th.
- Come by our table at the OSBC Open Business Conference on April 29-30th.
Wooohoooooooo!!!!!!!!!!!!!!!! This is the best news I have heard in ages! The new default engine for mysql!
Congratulations! Thats the best news ever!
what will happen to the mongodb stuff?
Thanks for the interest. We are looking for MongoDB experts to test our build on real-world workloads and benchmarks. Please email us at contact@Tokutek.com if interested in giving it a whirl.
The world is fractal!!! And you proved it;
for the community this is a fantastic gift that can lead to a new generation of relational applications on low cost hardware. Thanks
That’s great news. I am looking forward into delving into some code when I have some free time. Also easier to evaluate with clients as there no barrier to entry in getting the foot in the door.
Well done to the team.
Good news. I’ve downloaded and tested with my own application, the performance is about 1.2 – 2x compared to a vanilla mysql innodb.
Looking forward to the mongodb version.
I just read the section of PATENT CLAIMS.
“PATENT CLAIMS” means the claims of patents that are owned or
licensable by Tokutek, both currently or in the future; and that in
the absence of this license would be infringed by THIS
IMPLEMENTATION or by using or running THIS IMPLEMENTATION.
Does it mean that I have to apply the patent license from Tokutek if I run this implementation on my servers?
I’m not sure what you mean by “apply the patent license”. Can you clarify?
The patents (such as US Patent No. 8,185,551) are owned by and licensable by Tokutek.
Do I need to ask Tokutek to grant me a license for using the patents if I run TokuDB on my servers?
The section that you were reading about patents gives you the permission you need to run the software on your servers.
Was this ever tested to work with Percona XtraDB Cluster?
Enrico, not yet. We are at Percona/Live this week and have been discussing the required effort to enable TokuDB on Galera Cluster and Percona XtraDB Cluster.
Hello Tim,
are there any news from TokuDB + Galera Cluster? I would love to use it in our cluster.
We are considering a version of Galera that supports the TokuDB storage engine, stay tuned.
This page remains the top result for “tokudb galera” search result, so I’m putting this information here for completeness.
You can use galera & tokudb together using latest mariadb 10.0.x. See
https://downloads.mariadb.org/mariadb-galera/10.0.12/
https://mariadb.com/kb/en/mariadb/mariadb-documentation/mariadb-storage-engines/tokudb/how-to-enable-tokudb-in-mariadb/
https://mariadb.com/kb/en/mariadb/mariadb-documentation/replication-cluster-multi-master/galera/
http://stackoverflow.com/questions/22432633/php-pdo-mariadb-galera-cluster
Could you share a link that specifically mentions tokudb engine support in galera replication? I haven’t seen anything in the release notes on that. If it does working in galera I would assume it would be similar to myisam support which uses binary logging.