In this blog post, we’ll compare the performance of pattern matching queries vs. full-text indexes.

In my previous blog post, I looked for a solution on how we can search only a part of the email address and how can we make faster queries where the condition is email LIKE '%n.pierre%'. I showed two possible ways that could work. Of course, they had some pros and cons as well but were more efficient and faster than a like '%n.prierre%'.

But you could also ask why I would bother with this? Let’s add a FULLTEXT index, and everybody is happy! Are you sure about that? I’m not. Let’s investigate and test a bit. (We have some nice blog posts that explain how FULLTEXT …

In this post, we’ll look at how to implement ProxySQL with AWS Aurora.

Recently, there have been a few discussions and customer requests that focused on AWS Aurora and how to make the various architectures and solutions more flexible.

Flexible how, you may ask? Well, there are the usual expectations:

• How do you improve resource utilization?
• How can I filter (or block) things?
• Can I shard with Aurora?
• What is the best way to implement query caching?
• … and more.

The inclusion of ProxySQL solves many of the points above. We in Consulting design the solutions for our customers by applying the different functionalities to better match customers needs. Whenever we deal …

In this blog post, we’ll show that you can plot MySQL data in real time using Percona Monitoring and Management (PMM).

In my previous blog post, I showed how we could load into any metrics, benchmarks into MySQL and visualize it with PMM. But that’s not all! We can even visualize most any kind of data from MySQL in real time. I am falling in love with the MySQL plugin for Grafana — it just makes things so easy and smooth.

This graph shows us the number of visitors to a website in real …

In this blog post, we’ll look at how to migrate to Amazon RDS using Percona XtraBackup.

Until recently, there was only one way to migrate your data from an existing MySQL instance into a new RDS MySQL instance: take and restore a logical backup with mysqldump or mydumper. This can be slow and error-prone. When Amazon introduced Amazon Aurora MySQL, you could use Percona XtraBackup to take an online physical backup of your database and restore that into a new Aurora instance. This feature is now available for RDS MySQL as well. Using Percona XtraBackup instead of a logical backup can save a lot of time, especially …

In this blog post, we’ll look at the performance of multi-source replication with GTID.

Multi-Source Replication is a topology I’ve seen discussed recently, so I decided to look into how it performs with the different replication concepts. Multi-source replication use replication channels, which allow a slave to replicate from multiple masters. This is a great way to consolidate data that has been sharded for production or simplify the analytics process by using the same server. Since multiple masters are taking writes, care is needed to not overlook the slave. The traditional replication concept uses the binary log file name, and the position inside that file.

This was the standard until the release of global transaction identifiers (GTID). I have set up a test environment to validate which concept would perform better, and be a better choice for use in this topology.

SETUP

My test suite …

In this blog post, we’ll discuss some of the options available when purging binary logs. We’ll look at how to safely purge them when you have slaves in your topology and want to avoid deleting any binary log that still needs to be applied.

We generally want to ensure that, before purging the binary logs from the master, all logs were applied to the slaves to avoid halting them. The example error below is a classic case of a binary log purged before being applied on the slave:

Last_IO_Errno: 1236
Last_IO_Error: Got fatal error 1236 from master when reading data from binary log: ‘Could not open log file’

MySQL offers some options to purge of binary logs. One of them is executing the PURGE BINARY LOGS command. The  …

Please join Percona’s Principal Support Engineer, Sveta Smirnova, as she presents Effective Testing for Live Applications on March 29, 2018, at 10:00 am PDT (UTC-7) / 1:00 pm EDT (UTC-4).

When an application is hit with trouble in a live production environment, it is often difficult to:

• Repeat a problematic scenario without the risk of making things worse
• Find which query or sequence of actions caused the issue
• Prepare a dataset to share with the support team that they can use to investigate the problem

At the same time, it is not possible to solve troubles without understanding what caused them. This is why it is necessary to clearly …

In this post, I’ll discuss the fix for lp:1704195 (migrated to PS-2503), which prevents

ANALYZE TABLE

 from blocking all subsequent queries on the same table.

In November 2017, Percona released a fix for lp:1704195 (migrated to PS-2503), created by Laurynas Biveinis. The fix, included with Percona Server for MySQL since versions 5.6.38-83.0 and 5.7.20-18, stops

ANALYZE …

Please join Percona’s Principal Architect in Architecture & Projects, Yves Trudeau, as he presents ZFS with MySQL on Wednesday, March 28, 2018, at 7:00 am PDT (UTC -7) / 10:00 am EDT (UTC -4).

Are you curious about ZFS? Would you like to learn how to setup and configure ZFS? What about ZFS with MySQL?

ZFS on Linux has matured a lot. It offers unique features that are extremely compelling for use with a database server like MySQL.

During this webinar, we’ll review the main characteristics of ZFS, and walk through the configuration of ZFS and MySQL in order to provide good performance levels and superior ease-of-management. We will also cover aspects like backups using snapshots, cloning snapshots to create local slaves, the …

In this post, we’ll look at the MySQL 8.0 innodb_dedicated_server variable.

MySQL 8.0 introduces a new variable called innodb_dedicated_server. When enabled, it auto tunes innodb_buffer_pool_size, innodb_log_file_size and innodb_flush_method at startup (if these variables are not explicitly defined in my.cnf).

The new MySQL 8.0 variable automatically sizes the following variables based on the RAM size of the system:

innodb_buffer_pool_size:

• • <1G: 128M(default value if innodb_dedicated_server is OFF)
  • <=4G: Detected Physical RAM * 0.5
  • >4G: Detected Physical RAM * 0.75

innodb_log_file_size: …