Busy April 2017: MariaDB Dev Meeting (no-slave-left-behind, MyRocks, ...) and Percona Live

In a few days, I will start my yearly travel to North America which will bring me at Percona Live at the end of the month.  But I will first stop in New York to attend the MariaDB Developer Meeting.  Let's see what will happen there.

Percona Live

Booking.com is sponsoring the conference, and we will be present at the Monday Evening Reception.  You do not need a tutorial pass to attend: any valid passes will do.  If you do not have your ticket yet, it is time to register (you can use the discount code “SeeMeSpeak” for a 10% discount on the registration fees).

At the reception and during the conference, you will have to opportunity to discuss with our experts that are also speakers at Percona Live:

• Eric Herman - Principal Developer
• Simon Mudd - Senior Database Administrator
• Daniël van Eeden - Database Administrator
• Tarek Sheasha - Developer
• And me - Jean-François Gagné - Senior System Engineer

I will give more details about our talks in a follow-up post (by clicking on our name above, you can see those talks), but I can already write that those talks will be about RocksDB, MySQL Security, MySQL/MariaDB Replication, Scaling, Automation and Post-Mortem.

MariaDB Developer Meeting

On Sunday, April 9th and Monday, April 10th, the MariaDB Developer Meeting will take place in New York.  At this meeting, I will present How Booking.com avoids and deals with Replication Lag.  The goal of this presentation is to share knowledge and to define how we can do some modification to the No-Slave-Left-Behind MariaDB patch (MDEV-8112) to be used in a more generic way, maybe including in Galera and Group Replication.

I will also talk about the work Kristian Nielsen and I are doing to optimize replication when more than one storage engine is used.  In addition to InnoDB, TokuDB is already there and MyRocks is coming.  To provide replication crash safety without filesystem sync, it has been assumed that a single storage engine is used to store both data and replication position (mysql.gtid_slave_pos table).  This becomes more complicated with more than one storage engine, so Kristian and I are working on a way to make things work under the assumption that transactions do not spawn storage engine (a transaction is 100% InnoDB, 100% TokuDB or 100% MyRocks).  We have called this work Per-engine mysql.gtid_slave_pos and the progress on its implementation is tracked in MDEV-12179.

I am looking forward to seeing you at the MariaDB Developer Meeting and/or at Percona Live.