Galera was conceived from the ground up to be the simplest replication solution possible. And it really is very simple. But it is also very different from tradional replication approaches, and thus may be non-trivial to grasp at first sight.
If you looked at Galera replication for MySQL (Percona XtraDB Cluster, MariaDB Galera Cluster), even tried it, and asked one of the following... [read more]
Many strategies exist for increasing the performance of a database system. Some attempt to make each access to the data faster: for example, faster hardware may be used, or system settings may be tweaked. Other approaches attempt to decrease the total number of database operations: for example, by caching database queries or entire pages. Nevertheless, generally, performance decreases as the... [read more]
Many operations teams are tasked with competing goals of quickly delivering reliable infrastructure and also ensuring high uptime for all infrastructure components. By the time an organization has more than 50,000 hosts, this process is very well defined, usually the result of some very expensive mistakes made along the way. The purpose of this talk is to share best practices and lessons... [read more]
Facebook's compression improvements are included in official 5.6.7 release. In this talk, we are going to talk about the internal details of its implementation, related configuration parameters and the future improvements that are being worked on by Facebook.
InnoDB is the storage engine behind most busy MySQL databases in the world. How does it store data? How does it track free space? How does it retrieve rows? This presentation dissects InnoDB's storage formats and digs into the physical structure of tablespace files, extents, pages, records, indexes, and free space management. How many bytes would it take to store rows in a particular schema?... [read more]
Application PaaS or aPaaS is important for developers who want to build and run their applications in the cloud without having to deal with IaaS complexities directly. HP’s Converged Cloud strategy promises to help customers deliver and deploy services in the cloud leveraging open standards like OpenStack and a rich ecosystem of partner solutions. In this session you will become better... [read more]
You may have heard that there are performance benefits to using a column store instead of a row store. Column stores have become popular in closed source databases but they have recently become available in open source projects as well.
This talk will introduce the concept of column stores and contrast them with row stores. It will then introduce a set of open source column store... [read more]
ScaleDB delivers a clustered solution for virtualizing MySQL/MariaDB. It separates and virtualizes the database compute from the data itself; enabling CPU-bound databases to scale by adding compute nodes and I/O-bound databases to scale by adding storage nodes. Adding and removing nodes is done on the fly, without interrupting your application. ScaleDB will launch the product, and walk through... [read more]
Partitioning your database for the first time can be a rather stressful endeavor. This session provides a walkthrough of how to initially shard your MySQL infrastructure in a more pleasant way by leveraging the Jetpants toolkit. Open sourced by Tumblr in 2012, Jetpants provides easy-to-use automation for managing complex tasks in a partitioned MySQL environment.
The release of 5.6 is a pivotal moment in the history of MySQL, enabling both enhanced performance and impressive innovation. Yet what must be done to ensure MySQL stays competitive in the “world order of the cloud” and the growing reality of big data? Join panelists from HP, Continuent, Amazon Web Services, and Percona as they share their unique perspective about the reality and future of the... [read more]
It's not just about running a virtual machine with a NoSQL database on it; running databases in the Cloud has become a powerful option for many different types of workloads and scenarios.
In this keynote, the speaker will first examine what has happened so far in Cloud Computing from a Database perspective, and then explore future possibilities and trends.
Join Tomas Ulin, Oracle’s Vice President of MySQL Engineering, as he talks about the focus, strategy, investments and innovations evolving MySQL to power next generation Web, mobile, Cloud and embedded applications. Tomas will also discuss the latest and the most significant MySQL database release ever in its history, MySQL 5.6.
RecordingNoSQL solutions are oversold, but this is no reason for MySQL fans to become complacent. There are new challenges in data management, and relational databases must solve them or risk becoming irrelevant. This keynote will describe our work on Continuent Tungsten to deliver better infrastructure and architecture for MySQL-based systems. We will use our experiences to propose a roadmap for... [read more]
This year is the tenth anniversary of our gathering as a community to discuss and guide the direction of the MySQL ecosystem. With the release of MySQL 5.6, the increasing need for performance, the demand for big data support, and the question of the cloud, 2013 brings opportunity for an enormous leap in innovation. In his keynote, Peter Zaitsev gives his perspective on the state of the MySQL... [read more]
The future of the growth of the MySQL ecology and the driver of interesting technical innovation is in cloud computing. One key technology is improved cloud-oriented clustering. The existing public clouds operated by Amazon and by Google recognize the proven value of the MySQL ecosystem by offering MySQL as a managed database-as-a-service (DBaaS). Together, HP and Rackspace, as members of... [read more]
It is now over three years since Oracle acquired MySQL along with Sun Microsystems. Fears for the open source database's survival appear to have been misplaced. Oracle has increased investment in MySQL development, while a thriving ecosystem of potential alternatives and complementary product has emerged. This presentation covers the latest research from 451 Research into the state of the... [read more]
In this presentation, learn about the issues of managing a large number of instances of MySQL in the cloud, supporting 50 billion SQL statements per day. This presentation will cover the following topics using Amazon Web Services (EC2)
• What is AWS really costing you?
• Learn how to save with reserved and spot instances
• The need for monitoring and instrumentation
•... [read more]
Just as MySQL offers multiple backend storage engines Linux does similarly with a list of filesystem choices. Some are old, some are new, some have appealing features or options for tuning.
How do they play along with MySQL and its storage related goals? What considerations do you need to make when selecting the filesystem initially?
We consider RAID levels and settings,... [read more]
This talk is an introduction to load balancing between MySQL servers. The talk shows techniques which can be used in general, however we will focus on practical examples using HaProxy. The talk will cover the following topics.
- Load balancing a set of servers which are using master-slave replication.
- Load balancing in MySQL Cluster (ndbcluster).
- Load balancing in Percona... [read more]
The presentation illustrate an operative way for archiving the partition or tables as file, and how to restore them safely and quickly,
we will be presenting a case where historical data can be needed for future reuse, using simple file restore bypassing the need of reloading the data.
We also will help in getting a better understanding of what you can really do with DISCARD/IMPORT... [read more]
Setting up database resources on the cloud is something that lends itself well to being automated. One such solution to this automation is Chef. Chef is an open-source systems integration framework and is used for automating system deployment on the cloud. Learning chef is important to knowing how to take advantage of it's power, and is something especially important for DBAs to become... [read more]
What is MariaDB 10.0? Why did we jump version numbers? What new features are we including?
Also, a walk down memory lane covering features in MariaDB 5.5, 5.3, 5.2 & 5.1.
Expect new features about the new InnoDB, better error messages, performance schema, SHOW EXPLAIN, replication improvements including multi-source replication, and more.
Find out what we're doing... [read more]
The Cassandra Storage Engine allows access to data in a Cassandra cluster from MariaDB. Learn:
* How to make use of it
* When to use it, and if CQL makes sense
* Data & command mapping
* Use cases
* Future plans
And more. This will benefit MySQL users as well, since this is just a storage engine using the storage engine interface.
pt-query-digest is one of the more valuable components of the Percona Toolkit available as OSS to DBAs. In this talk, we will review the following with an eye towards real world test cases, output reviews and anecdotal production experience.
* Overview of functionality and usage
* Gathering data: slow/general logs, tcpdumps and more.
* How to use pt-query-digest against CSV... [read more]
In the third edition (2012) of High Performance MySQL, MHA usage is described as: "anyone other than Yoshinori who is using it in production, and we haven't used it ourselves." Little known to the authors, we've spent quite a lot of time working on MHA and figuring out its kinks. It has also been deployed in many a production environment.
Learn:
* how to get it running including... [read more]
A lot of people are interested in XtraDB Cluster as a replacement for conventional MySQL master/slave architectures to eliminate replication lag and finally achieve a highly-available masterless cluster of MySQL servers. But what would migration look like?
Drawing from a person-millienia of supervisory experience gained at Pythian, and in-depth research on human reliability in the fields of civil aviation, healthcare and nuclear safety, this session leads the audience through a compelling and vivid description of the stakes at hand in database operations, what we can learn from more mature fields of production engineering, and how to apply and... [read more]
In this talk we'll take a deep dive into the popular iostat and pt-diskstats tools and look at how they can be used to monitor performance on Linux. We'll look at the options associated with the tools to undestand how they influence behavior, the kernel metadata they both use as a basis for their calculations, and review the source code to understand the metrics that they report. We'll... [read more]
A monitoring system is arguably the most crucial system to have in place when administering and tweaking the performance of any database. DBAs also find themselves with a variety of monitoring systems and plugins to use; ranging from small scripts in cron to complex data collection systems. In this talk, I’ll discuss how Box made a shift from the Cacti monitoring system to OpenTSDB and the... [read more]
SymmetricDS is an open-source, flexible, extensible, asynchronous data replication engine specializing in the synchronization of multi-tier hierarchies of databases such as MySQL. SymmetricDS excels in applications where subsets of data need to be asynchronously replicated among hundreds or thousands of databases, as is the case in industries such as retail, telecommunications, medical,... [read more]
myq_gadgets is an open source toolkit of perl scripts I wrote long ago to help me more easily observe rates of change of global status variables in MySQL in an IO-stat style output.
This talk will introduce these tools and illustrate ways they can be helpful in monitoring and diagnosing MySQL issues with replication, Innodb, the query cache, application / database interaction, and... [read more]
MySQL 5.6 GA presented a lot of new features for both DBAs and developers, notable performance and scalability improvements, new security and monitoring features, and new query optimization methods. Most of them come with certain limitations and bugs though, and some of them requires DBAs to change their usual problem solving approaches. So it makes sense to discuss what DBAs and developers... [read more]
This session will be interesting for any person who cares about making MySQL fly! Many of recently published MySQL 5.6 benchmark results involved a lot of discussions around, and a lot of misunderstanding too.. So, the goal of this talk is to clarify all these points, including:
MySQL 5.6 includes the most significant enhancements and updates to MySQL replication ever released. It is a major step forward, in terms of performance, scalability, high availability and agility – whether deployed on premise or in the cloud.
Furthermore, MySQL 5.6 replication leverages DevOps methodologies, providing a compelling feature set coupled with a comprehensive array of... [read more]
The MySQL/InnoDB 5.6 release has many significant improvements and features. With the 5.6 release MySQL/InnoDB:
This will cover common tasks in RDS and gotchas for DBA's that are new to RDS.
1) Intro to RDS
a) security groups
b) parameter groups
c) setting things up correctly
2) Command line RDS tool
a) how to make administration faster and repeatable
3) Limitations of RDS and how to work around them.
a) slow query logs
b)... [read more]
Everyone knows that big data is here to stay, and one of the best NoSQL solutions out there is Hadoop. MySQL DBAs will need to know when to adjust and leverage Hadoop within their environment and even start using and administering Hadoop clusters. Hadoop is no replacement for the time-tested RDBMS, but for dealing with petabytes of data, Hadoop is the coolest kid on the block. In this session... [read more]
Abstract:
In this talk, I will discuss how to effectively use MySQL with KVM.
Virtualization has been there for long, however, it is quite underutilized and
also discourage in lieu of performance. However, for a minimal / no loss of
performance, one will be able to take advantage of high availability (obtained from
ability to move VMs across boxes) along with... [read more]
MySQL has come a long way in the past few years. But technology alone isn't sufficient. This required a good product (MySQL), good people and good process. We will explain how MySQL has remained relevant for Facebook. Bring your questions too as Facebook's MySQL performance team will be on hand to answer questions.
This will start of with a short presentation at the start and then a Q... [read more]
MySQL Cluster is perhaps the most badly named and most misunderstood product from the MySQL eco-system. Despite the name it is a completely separate product from the standard MySQL database. Due to the distributed nature of MySQL Cluster, it has a far more complicated architecture than a standard MySQL database. In this presentation, we will describe the architecture of the "MySQL Cluster"... [read more]
