Introduction

The purpose of this test is to benchmark MySQL 5.6 performance on hardware with Haswell CPUs, SSDs and PCIe Flash storage devices.

Background

Software

• SysBench OLTP workload installed on the database machine
  
• MySQL 5.6.24 distribution from Percona
  
• jemalloc used for MySQL Server and Sysbench test client
  
• Charts are plotted using MySQL Performance Analyzer

Method

• Data and software on server was wiped out post every test run.
  
• Predefined number of tables - 16 or 64
  
• Predefined size of rows - 20M per table
  

Tests
Read Only
Read Write
Concurrencies vary from 1 to 512 with incremental increase
EXT4 vs XFS

…

Capturing data is a critical part of performing a query analysis, or even just to have an idea of what’s going on inside the database.

There are several known ways to achieve this. For example:

• Enable the General Log
• Use the Slow Log with long_query_time = 0
• Capture packets that go to MySQL from the network stream using TCPDUMP 
• Use the pt-query-digest with the –processlist parameter

However, these methods can add significant overhead and might even have negative performance consequences, such as:

In two earlier posts, I gave some examples on how to use Perl to send tweets stored in a MySQL database to Twitter, and then how to automatically reply to your retweets with a “thanks”. In this post, I will show you how to automatically download your direct messages from Twitter, store the messages in a MySQL database, and then delete them.…

Last week, I was pleased to present at Percona Live Amsterdam 2015 regarding the importance of Performance Schema and how to begin approaching visualizing the data that is available to diagnose performance issues and fine tune your MySQL environment. You can download the slides from the Percona Live conference page.

The session highlighted using the ELK (Elasticsearch+Logstash+Kibana) stack to assist visualizing event data, in addition to graphite+graphana to visualize time-series data.

As many people who attend conferences are aware, the availability of internet access is sketchy at best. To combat this, the visualization stack that I created was strictly local utilizing Docker, specifically the Docker Toolbox.

The docker-compose.yml …

Indexes are separate data structures that provide alternate pathways to finding data. They can and do generally speed up the processing of queries and other DML commands, like the INSERT, UPDATE, REPLACE INTO, and DELETE statements. Indexes are also called fast access paths.

In the scope of the InnoDB Database Engine, the MySQL database maintains the integrity of indexes after you create them. The upside of indexes is that they can improve SQL statement performance. The downside is that they impose overhead on every INSERT, UPDATE, REPLACE INTO, and DELETE statement, because the database maintains them by inserting, updating, or deleting items for each related change in the tables that the indexes support.

Indexes have two key properties—usability and visibility. Indexes are both usable and visible by default. That means they …

On Percona Live! Amsterdam 2015 we had a talk with Peter Boros about GTID replication.

Here are the slides.

Running gtid replication in production from Balazs Pocze

Well, it was ended a week ago, but I had too many errands to run, so I couldn’t post anything about it.

It was really great, again.This was the third time I attended (2013 London, 2015 Santa Clara) so now I have met with a lot of familiar people – it is true that MySQL has a great community. The chosen city was great, Amsterdam is one of the coolest places of Europe, the hotel was neat, and the programs were also astounding.

The conference sessions were great too, I really enjoyed, them all, and because they are running on 8 thread parallel it is not that bad that there are some recurring sessions; if you missed one in spring you can watch it on autumn.

So, everything was comfy and neat. I hope I’ll attend on the next one too …

There were a few topics where I plan to dig deeper in the next weeks

• ProxySQL because HAProxy is a good choice, but it is only speaks TCP and HTTP but not …

Hi!

I'm speaking at Oracle Open World this October 29th (Thursday). My Session is a hands-on lab session: HOL3348 on MySQL Cluster.

If you are interested in familiarize a bit with MySQL Cluster this is definitely a session for you.

I will start by briefly introducing MySQL Cluster and its Architecture.

Then I will guide you through the needed steps to:

• Install a local MySQL Cluster
• Start MySQL Cluster
• Connect to MySQL Cluster (using the command line)
• Ways to monitor MySQL Cluster
• Safe shutdown of MySQL Cluster

We will have a chance to see which are the most common commands and operations used in MySQL Cluster Administration.

If there is …

MySQL 5.7 comes with a new backup tool, named mysqlpump, which is almost the same as mysqldump with the ability of extracting data in parallel threads.

I tried a little experiment. Using a server containing 11 databases, with a total of 300 tables and about 20 million rows (roughly ≈ 10GB,) I used both mysqldump and mysqlpump to get a backup.

mysqldump --all-databases  > dump.sql

mysqlpump --all-databases \
    --add-drop-database --add-drop-table --skip-watch-progress \
    --default-parallelism=10 \
    --parallel-schemas=db,db1,db2 \
    --parallel-schemas=db3,db4,db5 \
    --parallel-schemas=db6,db7,db8 \
    --parallel-schemas=db9,db10 > pump.sql

The backup with mysqldump took 3 minutes and 33 seconds. The one with mysqlpump took 2 minutes and 55 …

Last week, 21-23 September, it took place the European MySQL Conference, or “Data performance Conference” as this year’s subtitle was “MySQL. NoSQL. Data in the cloud.”. This year, it changed its location from London to Amsterdam and, as most people I talked to agreed, the change was for good. As every year, Percona was the company organizing it, but it had the participation of all the major players in the open source MySQL/MongoDB/Cloud data world. Special mention goes to Booking.com, which had more …