Percona Live: Data Performance Conference 2016 Logo

April 18-21, 2016

Santa Clara, California

Schedule - Day 3 | Thursday 21 April 2016

Thursday 21 April 2016
8:15am to 9:00am
9:00am to 9:25am
  • Sponsored
Data is critical to the success of your business. You rely on your database and its supporting systems to power the applications that drive your business. These systems must be reliable, scalable, efficient – and increasingly, open source.
Other
9:30am to 9:50am
  • Sponsored
As database engineers, we have had to support the crazy dreams of application developers since the beginning of the internet. I say let’s take back the power! It’s time to use a tool that was built for just this problem. I’m talking about Apache Cassandra and how it can help you eliminate down time or span your data around the world with ease.
Other
9:55am to 10:10am
10:15am to 11:00am

11:00am to 11:50am

11:00am to 11:50am
  • New and Trending Topics
We always say Galera replication is synchronous, but is it really? Always? For every step? What are the caveats of such replication? What does virtual-synchronous mean? In this talk, I'll discuss what certification and group communication are, and how they're performed. I'll also cover the difference between MySQL 5.6 GTID and Galera GTID.
MySQL Other
11:00am to 11:50am
  • Architecture/Design
  • New and Trending Topics
Are you a DBA/devops tired of being asked to also support MongoDB? Wouldn't it be cool if your current RDBMS could support MongoDB...
MySQL MongoDB Java
11:00am to 11:50am
  • Development
  • Scalability/Performance
There are substantial improvements in optimizer in MySQL 5.7. Most noticeably, users can now combine relational data with noSQL. A new JSON datatype has been introduced. Users can easily search and manipulate JSON documents with the help of built-in JSON functions. Functional indexes are now supported through both stored and virtual generated columns.
MySQL Optimization NoSQL
11:00am to 11:50am
  • Architecture/Design
  • Scalability/Performance
,
Manhattan is a high performance, multi-tenant, eventually consistent KV store developed at Twitter. Manhattan comes with native support for LSM backend. Its architecture supports multiple backend solutions. For use cases where a btree backend provides more scalable and predictable performance we have developed a solution to use InnoDB as the storage engine of choice.
MySQL InnoDB NoSQL
11:00am to 11:50am
  • Scalability/Performance
In MySQL, sharding is one of the most used (and most abused!) MySQL scaling technology. Before employing this technique, it’s important to assess whether sharding is right for your application, or would another scaling technology work better? What specific technologies can assist you with your sharding implementation? Have you answered all the pertinent questions before you start sharding?
MySQL Sharding
11:00am to 11:50am
  • Architecture/Design
  • Scalability/Performance
In database technology the law of the instrument is a bit overused these days but for a reason! Yelp, like a lot of companies which have been around for a while, has a set of tools we use because we are happy with them. For the Database Engineering team, the tool of choice is MySQL, and it’s a mighty hammer.
MySQL Cassandra NoSQL
11:00am to 11:50am
  • Case Stories
  • Operations and Management
In this session I would like to show you how we automate all the operations-related work at Gawker Media.
Ansible Puppet Devops
11:00am to 11:50am
  • Operations and Management
  • Scalability/Performance
,
MySQL, like any other production service, needs to be monitored whether it be for potential performance problems, availability issues, or utilization as part of capacity planning. But with so many metrics exposed by our operating systems and MySQL, what exactly should one monitor and alert on?
MySQL Tools and Techniques Monitoring
11:00am to 11:50am
  • Scalability/Performance
Working primarily with PostgreSQL, I've noticed, that an open source database DBA must be a Linux/UNIX expert. PostgreSQL, MySQL or any NoSQL database challenges an administrator with a lot of operating system tasks in comparison with our colleagues from commercial databases.
MySQL Other Optimization
11:50am to 12:50pm

12:50pm to 1:40pm

12:50pm to 1:40pm
  • Case Stories
  • New and Trending Topics
In the last 24 months, MySQL replication speed has improved a lot thanks to implementing parallel replication. MySQL and MariaDB have different types of parallel replication; in this talk, I present in details the different implementations, with their limitations and the corresponding tuning parameters. I also present benchmark results from real Booking.com workloads.
MySQL Tools and Techniques Replication
12:50pm to 1:40pm
  • Architecture/Design
  • New and Trending Topics
,
MongoDB adds Pluggable Storage Engine capability. This exciting new feature can been exploited to provide high performance, large capacity storage capabilities to your MongoDB applications. Learn how these new storage engine technologies can enhance your application.
MongoDB Optimization NoSQL
12:50pm to 1:40pm
  • Operations and Management
, ,
Shared locks, exclusive locks, intention locks, gap locks, deadlocks, and lock wait timeouts! How does it really work under the hood? And how can we use this knowledge to make our environment better? This presentation will be a deep dive into InnoDB locking internals, which will help attendees be able to make strategic technology decisions (i.e. is InnoDB the right choice?).
MySQL InnoDB Fundamentals
12:50pm to 1:40pm
  • New and Trending Topics
,
Amazon, already very successful with its RDS database as a service platform, a few months ago started a new database service: Aurora. What's different about Aurora? Which tasks are it suitable for it? What kind of load is bad for it? In this talk we will review what is known about the Aurora architecture.
MySQL AWS Data as a Service
12:50pm to 1:40pm
  • Case Stories
, ,
The MySQL Utilities from Oracle have been available for some time but how useful are they? In this talk we will cover the various tools available and how we are using then in our shops to reduce our work load , increase efficiency and ensure data consistency. We are all familiar with Percona Toolkit but the MySQL Utilities seem not to be as well known or visible.
MySQL Devops Tools and Techniques
12:50pm to 1:40pm
  • Architecture/Design
  • Big Data
,
Seznam.cz is the largest and the most visited web portal and search engine in the Czech Republic. It is one of a few search engines in the World which successfully competes with Google in the field of local full-text search. Besides the Search engine, Seznam runs over 40 different web services such as News portals, Map portal, Email service and many more.
MySQL Hadoop Business Intelligence
12:50pm to 1:40pm
  • Operations and Management
Many MySQL database administrators start out as software developers, sysadmins or devops engineers who are assigned to handle the databases without any formal training or mentoring. They then rely on Google to answer any questions.
MySQL Fundamentals
12:50pm to 1:40pm
  • Operations and Management
In a monitor/measure everything environment sometimes "Less is More!" to get the big picture.
MySQL Metrics Monitoring
12:50pm to 1:40pm
  • Scalability/Performance
Systems performance provides a different perspective for analysis and tuning, and can help you find performance wins for your databases, applications, and the kernel. However, most of us are not performance or kernel engineers, and have limited time to study this topic.
Other Optimization Tools and Techniques

1:50pm to 2:40pm

1:50pm to 2:40pm
  • New and Trending Topics
  • Scalability/Performance
,
Most global companies gather data from several sources, centralize it and then query it for analytics/reporting using an array of interfaces. Before MariaDB 10.0 and MySQL 5.7, replication from multi-sources was possible by applying remote binary logs locally and/or doing workarounds with some other kind of development.
MySQL Tools and Techniques Replication
1:50pm to 2:40pm
  • Security
Data security is a concern for most people, and recent reports in the news show that significant amounts of user details stored in MongoDB are easily accessible. Does it mean that MongoDB is not able to secure your data?
MongoDB Fundamentals NoSQL
1:50pm to 2:40pm
  • Scalability/Performance
,
In this session we will discuss selected areas of InnoDB and XtraDB 5.7 internals that are mostly related to buffer pool management, from a performance and scalability point of view.
MySQL InnoDB Programming
1:50pm to 2:40pm
  • Case Stories
  • Scalability/Performance
"It looks like a database problem" This is a pretty common phrase that comes up when triaging a production issue if MySQL is in the critical path.
MySQL Devops Tools and Techniques
1:50pm to 2:40pm
  • Operations and Management
Everybody who’s dealt with MySQL in production knows about innodb_buffer_pool; what about the other configuration variables that have bad defaults that will cost you time and money? Learn why sql_mode is your engineering team’s best friend, why if you haven’t set the multiple charset variables properly you’re probably in trouble, and more.
MySQL Devops
1:50pm to 2:40pm
  • Sponsored
, ,
Following the Introduction to MySQL Document Store session, we give a more in-depth tour of its implementation and core components.
Other
1:50pm to 2:40pm
  • New and Trending Topics
Shard-Query is a analytics horizontal scale-out layer for MySQL that is comparable in many aspects to RedShift, and can in fact, outperform RedShift for the same price/performance ratio. This talk introduces the technology of Shard-Query, compares to the technology of RedShift, and offers a comprehensive performance comparison using the Star Schema Benchmark.
MySQL Sharding Other
1:50pm to 2:40pm
  • Analytics
  • Operations and Management
The Prometheus timeseries monitoring system is a perfect match for collecting detailed metrics from MySQL. Prometheus can provide both performance analysis and health alerting.
MySQL Metrics Monitoring
1:50pm to 2:40pm
  • Architecture/Design
  • Development
Several years ago I had a task to develop custom storage for IM history. Started from simple index tree over mmap(2), I quickly found that mmap(2), madvise(2) & Co. do poor job to be used as the foundation for database engine. So I switched to direct IO and started to develop my own memory management, advanced eviction, IO scheduling with adaptive read ahead and surely transaction processing.
Programming Fundamentals

3:00pm to 3:50pm

3:00pm to 3:50pm
  • Case Stories
  • Operations and Management
,
Verisure, Europe's most popular home alarm with more than 1.8 million customers using the system every day, relies on MySQL for a lot of its customers' data. Data is stored into a sharded database architecture using MySQL.
MySQL Tools and Techniques Business Intelligence
3:00pm to 3:50pm
  • Operations and Management
MongoDB is one of the most popular NoSQL solutions and as MongoDB is Open Source, any developer can contribute to the product and anyone can use it. The business model of the company behind MongoDB is to sell premium support and administration tools to maintain and monitor MongoDB. As an alternative there are (open source) solutions that could make your life as a DBA easier.
MongoDB Devops Monitoring
3:00pm to 3:50pm
  • Architecture/Design
  • Scalability/Performance
This session will cover most hot performance issues in MySQL 5.5, 5.6 and 5.7 related to scalability and high loads, bringing on the table some initial design limitations in MySQL and InnoDB, and their step-by-step improvements, remaining issues, workarounds, final solutions, and many other stuff around (including OS, FS, low level HW issues, storage impact, etc.) And, of course, all of presented
MySQL InnoDB Tools and Techniques
3:00pm to 3:50pm
  • Analytics
  • Big Data
Cluster computing frameworks such as Hadoop or Spark are tremendously beneficial in processing and deriving insights from data. However, long query latencies make these frameworks sub-optimal choices to power interactive applications.
Java NoSQL Business Intelligence
3:00pm to 3:50pm
  • Case Stories
  • Operations and Management
, , , ,
Ever wonder how DBAs function at other companies? Curious about “the right way” to manage an on-call schedule?
MySQL Devops
3:00pm to 3:50pm
  • Analytics
  • New and Trending Topics
Time series is a fascinating subject that is becoming more and more interesting and relevant for all the DBAs and Data Architects.
MySQL Metrics Internet of Things
3:00pm to 3:50pm
  • Operations and Management
, ,
Managing a large farm is often difficult and tedious, especially with variances in hardware and configurations among different machines. Orchestrator makes managing a large set of MySQL instances easy with the distribution of agents to every host, and a central orchestrator to issue commands.
MySQL Devops
3:00pm to 3:50pm
  • Operations and Management
Containers are revolutionizing the way we deploy and maintain our infrastructures: reducing development overhead, streamlining dev / test / ops, and enabling highly scalable, dynamic infrastructures. But containers still have a key problem: monitoring and troubleshooting them is impractical, painful, and sometimes plain impossible.
Devops Monitoring
3:00pm to 3:50pm
  • Operations and Management
  • Scalability/Performance
Linux has a long and successful history, but as a result of the 25 years of its existence there are an increasingly large number of kernel parameters, filesystems heuristics, and other things which make it perform poorly for modern database workloads. From the TCP stack to things like NUMA, running high-performance Linux systems is fraught with peril.
MySQL Devops Optimization