Percona Live: Data Performance Conference 2016 Logo

April 18-21, 2016

Santa Clara, California

Schedule - Day 1 | Tuesday 19 April 2016

Tuesday 19 April 2016
8:00am to 9:00am
None
9:00am to 9:15am
None
  • Sponsored
Peter Zaitsev (Percona)
Peter Zaitsev welcomes everyone to Percona Live Data Performance Conference 2016
Other
9:20am to 9:45am
None
  • Sponsored
Chad Jones (Deep Information Sciences)
Machine learning enables businesses to gain competitive advantage through predictive analytics. But if we look deeper into the data stack, we find the need for the same predictive capabilities for MySQL tuning. With over 1013 possible tuning permutations, some requiring reboots or a rebuild, DBAs spend too much time on MySQL tuning for a point-in-time situation that changes constantly. This session will show how unsupervised machine learning based on resource, workload and information modeling can predictively and continuously tune databases.
Other
9:50am to 10:50am
None
  • Sponsored
Bill Nye (Planetary Society)
The best science is built on solid data. As a world-renown icon in tech and geek circles everywhere, Bill Nye fights to raise awareness of the value of science, critical thinking, and reason. He hopes that the data he brings will help inspire people everywhere to change the world. Bill is the CEO of the Planetary Society.
Other
10:50am to 11:30am
None

11:30am to 12:20pm

11:30am to 12:20pm
Ballroom A
  • Operations and Management
  • Scalability/Performance
Luis Soares (Oracle)
The most recent MySQL release comes with a very exciting set of replication features. They make replication more operations friendly, further reduce the administrative overhead, provide a much improved user experience, and enable running MySQL smoothly both at scale and/or in very tightly coupled network environments such as in a cluster.
MySQL, High Availability, Replication
11:30am to 12:20pm
Ballroom B
  • Architecture/Design
  • Scalability/Performance
Bob Burgess (Shopify)
Different patterns of data (or data access) in MySQL can often benefit from out-of-the-ordinary methods of storing or arranging the data. The main choices include storage engines, storage hardware, and table structure. Combinations of these three variables alone can be overwhelming. In this talk, I will discuss several shapes of data and data access and reveal results of testing various combinations of engine, disk, and table structure for them. Several of these tests stemmed from requirements in our environment.
MySQL, InnoDB, Optimization
11:30am to 12:20pm
Ballroom C
  • Operations and Management
  • Scalability/Performance
Leif Walsh (Two Sigma)
Whether your data's in MySQL, a NoSQL, Hadoop, or somewhere in the cloud, you're likely paying decent money for storage and IOPS. With ever-growing data volumes, and the need for SSDs to cut latency and replication to provide insurance, your storage footprint is an important place to look for savings.
MySQL, MongoDB, Metrics
11:30am to 12:20pm
Ballroom D
  • Analytics
  • Big Data
Torkel Ödegaard (Raintank INC)
An introduction to the open source software Grafana, a graph and dashboard composer with rich metric query builders and visualizations. Learn why Grafana has quickly become the leading frontend for visualizing time series metrics. Grafana supports time series databases like Graphite, InfluxDB, OpenTSDB and Elasticsearch. A demo rich session that will also show the latest features, use cases and tips & tricks.
Metrics, Monitoring, Business Intelligence
11:30am to 12:20pm
Ballroom E
  • New and Trending Topics
  • Scalability/Performance
Vadim Tkachenko (Percona)
TokuDB is write-optimized storage engine and naturally it is quite different from traditional B-Tree-based engines. I will talk about what is different about it, and how it should be used and configured to get the best experience. TokuDB is a specialized engine: we can't expect it to be a silver bullet for all cases! We will discuss when it shines -- and when it does not. Internal details and benchmarks are included!
MySQL, Optimization
11:30am to 12:20pm
Ballroom F
  • Sponsored
Alfredo Kojima (Oracle), Jan Kneschke (Oracle), Mike Frank (Oracle)
MySQL 5.7.12 includes a new Document Store feature for MySQL. The MySQL 5.7 JSON support enables an entirely new category of use cases for managing data. The Document Store functionality expands on that with a Server plugin, a new API and a suite of new components designed to make MySQL accessible for users who are not familiar with the SQL language or prefer a schemaless data store. This session gives a broad, high level introduction as to what the Document Store is about, its client components, what you can do with it, why you'd want to and how.
Other
11:30am to 12:20pm
Ballroom G
  • New and Trending Topics
  • Operations and Management
David Murphy (Percona)
MongoDB's Enterprise licence adds many features. However, getting their Ops Manager feature is still a pricey option.Ops Manager is an expensive cloud-based platform for alerts, monitoring, schema design, and deployment. And this all comes with vendor lock in. In this session, you will learn about: - Alternative cloud-based vendors - Public provisioning models like Ansible and Chef - Visualizing the schema - Monitoring on your own terms - Building a backup manager
MongoDB, Devops, Monitoring
11:30am to 12:20pm
Ballroom H
  • Analytics
  • Development
Riddhi Shah (Rackspace), Timothy Given (Rackspace), Steve Croce (Rackspace)
Offering HA Redis and MySQL as a service on the same platform produces great benefits and a series of challenges from infrastructure to the user interface. Members of the Rackspace Cloud Databases and ObjectRocket Redis product teams will discuss their architectural choices and the challenges they faced in development.
Ruby, Optimization, Monitoring
11:30am to 12:20pm
Room 203
  • Scalability/Performance
Ken LeTourneau (Intel)
This talk looks at the performance of Percona Server for MySQL for Linux running on the same Intel system but with three different storage configurations. We will look at and compare the performance of (1) a RAID of HDD vs. (2) a RAID of SATA SSD vs. (3) a RAID of NVMe SSD. In the talk we’ll cover the hardware and system configuration and then discuss results of TPC-C and TPC-H benchmarks. Then we’ll look at the overall system costs including hardware and software, and cost per transaction/query based on overall costs and benchmark results.
Tools and Techniques, Data in the Cloud
11:30am to 12:20pm
Room 204
  • MySQL 101
  • Operations and Management
Michael Benshoof (Percona)
The goal of this talk is to review the key features that would be part of a schema review, including things such as largest tables, index issues and table activity.
MySQL
11:30am to 12:20pm
Room 209
  • Big Data
  • Case Stories
Demeter Sztanko (Badoo), Alexey Eremihin (Badoo)
Badoo is a dating-focused social networking service. The site operates in 180 countries and is most popular in Latin America, Spain, Italy and France. Badoo is being used by tens of million users on a monthly basis. Business Intelligence department is receiving, processing and storing around 200 000 events per second from around 150 various data sources. We will be talking about the technology behind Business Intelligence at Badoo, the challenges we face, it's evolution, success stories and our failures.
Hadoop, Spark, Business Intelligence
11:30am to 12:20pm
Room 210
  • Development
  • Scalability/Performance
zhai weixiang (Alibaba)
The MySQL Binlog Replication has a lot of advantages. For example, the binary log is more readable and can be used for various scenarios. And we can even build a very complicated replication topology with careful design; But on the other hand, it requires more disk written, two fsync operations to make a transaction durable. The traditional replication mechanism always constitute a serious bottleneck that limit the ability of scaling up the write workload. In this session , I’d like to introduce another replication mechanism: physical replication based on InnoDB.
MySQL, InnoDB, Replication
12:20pm to 1:20pm
None

-

1:20pm to 2:10pm

1:20pm to 2:10pm
Ballroom A
  • Architecture/Design
  • Operations and Management
Simon Martin (Facebook)
MySQL semi-synchronous replication is a powerful feature allowing very fast failover with no data loss. However managing it operationally is tricky as you move in to an environment where replication stopping will cause writes to stop too.
MySQL, High Availability, Replication
1:20pm to 2:10pm
Ballroom B
  • Architecture/Design
  • Scalability/Performance
Kimberly Wilkins (ObjectRocket by Rackspace)
With the changes to MongoDB since 2.6 and even more with 3.x, the decision to shard has been made a little more complex. In this presentation Kimberly will discuss the implications of recent MongoDB changes and how those come into play with the decision to shard or use one large replica set. She will review the different storage engines, their impacts on sharding and on scaling, and talk about the pros and cons for each engine with differing workloads.
MongoDB, NoSQL, Sharding
1:20pm to 2:10pm
Ballroom C
  • Development
  • Scalability/Performance
Olav Sandstå (Oracle)
This presentation gives an introduction to the MySQL query optimizer. Knowing how queries are optimized and executed will help you write better performing queries. The main phases of the MySQL optimizer and the main optimization strategies are presented. Examples of optimization strategies that will be presented are query transformations, data access strategies, the range optimizer, the join optimizer, and subquery optimization. An overview of the cost model and how the optimizer does cost estimations are included.
MySQL, Optimization, Programming
1:20pm to 2:10pm
Ballroom D
  • Architecture/Design
Alexey Maykov (MemSQL)
MyRocks is a new open source storage engine for MySQL. It is jointly developed by engineers from Facebook and MariaDB. The engine is built on top of RocksDB storage library. Since it uses LSM as a data format, it provides much better compression ratio and much smaller write rate. It is especially beneficial to use MyRocks engine for write heavy workloads on flash storage. In this talk, I will talk briefly about characteristics and benefits of MyRocks. I will spend most of the time looking at internals. I will cover the structure of LSM tree, tuning and parameters.
MySQL, Devops, Tools and Techniques
1:20pm to 2:10pm
Ballroom E
  • New and Trending Topics
  • Operations and Management
Brandon Bercovich (Uber)
Companies often have to build their infrastructure to handle the peak load of their systems while being over-provisioned during off-peak times. At Uber we are leveraging Apache Mesos, Apache Aurora, and other frameworks to make our systems more efficient, reliability and to take advantage of excess off-peak capacity for use in resource-intensive jobs like Hadoop. We will go over why we are using Apache Mesos, the pros and cons of the frameworks we are using, and the benefits and challenges of running a multi-framework environment to handle the various services and jobs we have.
Hadoop, Spark, Devops
1:20pm to 2:10pm
Ballroom F
  • Big Data
Patrick McFadin (DataStax)
You would love to try some Apache Cassandra but that relational brain is still on? You aren’t alone. Let’s take that OLAP data model and get it OLTP. As you will see, static schema, one-to-many, many-to-many still have a place in Cassandra. From the familiar, we'll go into the specific differences in Cassandra and tricks to make your application fast and resilient. Real techniques to translate application patterns into effective models and common pitfalls that can slow you down and send you running back to RDBMS land.
Cassandra, Tools and Techniques, NoSQL
1:20pm to 2:10pm
Ballroom G
  • Operations and Management
Aleksandr Kuzminsky (Box, TwinDB)
Even without available backups InnoDB's design allows to recover database after many kinds of accidents. The session will cover: * InnoDB files and their formats * Recovery corrupted databases * Recovery after DROP TABLE * Recovery compressed InnoDB tables * Table structure recovery from FRM files and InnoDB dictionary
MySQL, InnoDB, Tools and Techniques
1:20pm to 2:10pm
Ballroom H
  • Scalability/Performance
Dave Anselmi (Clustrix)
Are you tired of supporting complicated MySQL installations? Arrays of lagging Read Slaves? Complicated Multi-Master replication without significant write-scale? Increasingly fragile Sharding topologies that get harder and harder to support? What if MySQL could scale-out like web servers and NoSQL databases? Join us to learn about a MySQL compatible database which actually Scales-Out: Each new database node adds linear write and read scale, providing 10x the performance of MySQL without modifying your enterprise applications.
MySQL, Sharding, Replication
1:20pm to 2:10pm
Room 203
  • Scalability/Performance
Kyle Bader (Red Hat), Yves Trudeau (Percona)
Many operators have selected OpenStack as their control plane of choice for providing both internal and external IT services. Year after year, the OpenStack user survey has shown Ceph to be the dominant backend for providing persistent storage volumes through OpenStack Cinder. As developers build new applications, and repatriate old workloads, they are quickly discovering the need to provide database services within their OpenStack infrastructure.
MySQL, OpenStack, Docker
1:20pm to 2:10pm
Room 204
  • MySQL 101
  • Operations and Management
Michael Benshoof (Percona), Sveta Smirnova (Percona), Francisco Bordenave (Percona)
The question comes in from your primary stakeholders: "The application is slower than normal so it must be the database." Where do you start? Troubleshooting performance issues can be a bit tricky if you don't know what to look for and how to look at it. There are a lot of moving pieces that interact within a MySQL instance that may need your attention. In this session we will show you how we do it at Percona, what to monitor and track, what to check and how so you will learn about our secret sauce to improve MySQL performance.
MySQL
1:20pm to 2:10pm
Room 209
  • New and Trending Topics
  • Scalability/Performance
Nikita Ivanov (GridGain Systems)
This talk will provide deep introduction to one of the fastest growing Apache project: Apache Ignite - In-Memory Data Fabric. Apache Ignite provides highly integrated software defined in-memory layer that support traditional In-Memory Data Grid and Compute Grid functionality, in-memory straming and file system, MPP/MPI style high-performance computing as well as acceleration for Apache Hadoop for storage for Apache Spark.
Java, Data as a Service, High Availability
1:20pm to 2:10pm
Room 210
  • Scalability/Performance
Sunny Bains (Oracle)
Find out what's new in the InnoDB 5.7 release * Scalability improvements * Virtual columns * Functional indexes * Improved ALTER TABLE performance * Transparent Compression * Changes to internals * Improved flushing An opportunity to understand how to use these new features and improvements.
MySQL, InnoDB, Fundamentals

2:20pm to 3:10pm

2:20pm to 3:10pm
Ballroom A
  • Case Stories
David Turner (Uber), René Cannaò (Self Employed)
This talk explains why Dropbox considered rolling out global transaction IDs so early.
MySQL, Devops, Tools and Techniques
2:20pm to 3:10pm
Ballroom B
  • Scalability/Performance
Jason Terpko (ObjectRocket by Rackspace)
MongoDB Chunks – Distribution, Splitting, and Merging This presentation will discuss how the distribution of chunks can create scenarios where you may need to manually move, split, or merge chunks in your sharded cluster. Scenarios requiring these actions can exist with both optimal and sub-optimal shard keys. This can be especially true when a new change has been introduced to the workload. It may cause hot-spotting, unbalanced storage, or empty chunks.
MongoDB, NoSQL, Sharding
2:20pm to 3:10pm
Ballroom C
  • Scalability/Performance
Nicolai Plum (Booking.com)
This talk will present the different dimensions in which attendees will need to scale up their MySQL installations as their enterprises grow. First we will discuss each dimension of scalability with details and examples. We will then survey methods of scaling the MySQL-based technology stack, covering techniques of systems engineering and technologies available to solve the problems enterprises face today. Scalability directions include challenges of data size, complexity, and query rates.
MySQL, Tools and Techniques, Fundamentals
2:20pm to 3:10pm
Ballroom D
  • Architecture/Design
  • Case Stories
David MacKenzie (Box)
Box's /events API powers our desktop sync experience and provides users with a realtime, guaranteed-delivery event stream reflecting all of the activity happening inside their accounts. To support this functionality, we chose to leverage HBase to store and serve a separate message queue for each of our 30+ million users.
Hadoop, Replication
2:20pm to 3:10pm
Ballroom E
  • Case Stories
  • Scalability/Performance
Ernie Souhrada (Pinterest)
In the age of social networking, mobile apps, and the unstoppable panopticon of government surveillance, the SQL verbs "DELETE" and "TRUNCATE" often feel like forbidden four-letter words. Between our phones and tablets, computers and cars, and countless other "smart" devices, exabytes of content and clickstreams are generated every day.
MySQL, InnoDB, Optimization
2:20pm to 3:10pm
Ballroom F
  • Architecture/Design
  • Scalability/Performance
Avi Kivity (ScyllaDB)
Scylla is a new NoSQL database that applies systems programming techniques to a horizontally scalable NoSQL design to achieve extreme performance improvements. Scylla is capable of exceeding one million requests per second per node, while providing Apache Cassandra compatibility, scaling, and geographical distribution properties. The Scylla design is based on a modern shared-nothing approach. The Scylla design is based on a modern shared-nothing approach.
Cassandra, Optimization, High Availability
2:20pm to 3:10pm
Ballroom G
  • Development
  • Operations and Management
Jenni Snyder (Yelp)
You're probably already using automation to build your application, manage configuration, and alert you in case of emergencies. What's keeping you from doing the same with your MySQL schema changes? For us, the answer was "lots of things". Today, Yelp uses Liquibase to manage our schema changes, pt-online-schema-change to execute them, and Jenkins to ensure that they're run in all of our environments. During this session, I'll explain the history of MySQL schema management at Yelp, and how hard it was for both developers and DBAs.
Other, Devops, Tools and Techniques
2:20pm to 3:10pm
Ballroom H
  • Scalability/Performance
Chad Jones (Deep Information Sciences), John Teague (GEMServers LLC)
Relational databases have a debilitating flaw: No matter how rigorously you stay on top of tuning, they slow to a crawl when hit with large-scale data and traffic conditions—driving you crazy while putting your applications and business at risk. GEMServers, a white-glove WordPress hoster whose business depends on delivering high performance at scale, couldn’t afford that risk. GEMServers’ founder will describe his quest to find a solution, and why and how application-aware, adaptive MySQL delivers high-velocity data streaming alongside real-time analytics, at scale.
MySQL
2:20pm to 3:10pm
Room 203
  • New and Trending Topics
  • Scalability/Performance
Yves Trudeau (Percona), Kyle Bader (Red Hat), Brent Compton (Red Hat)
In this session, we’ll compare the performance of MySQL on public and private clouds, with a head-to-head look at: (a) MySQL on Amazon AWS EBS (b) MySQL on Amazon AWS EBS Provisioned IOPS (c) MySQL on an OpenStack/Ceph private cloud (SuperMicro HDD-based Ceph storage) (d) MySQL on an OpenStack/Ceph private cloud (SuperMicro all-flash Ceph storage) (e) MySQL on a single bare metal SuperMicro server (baseline) Thanks to SuperMicro, Seagate, Samsung and Mellanox for co-sponsoring the server and flash lab.
MySQL, OpenStack, Data in the Cloud
2:20pm to 3:10pm
Room 204
  • Architecture/Design
  • MySQL 101
Jervin Real (Percona), Michael Coburn (Percona)
Preventing downtime-causing emergencies in MySQL can be difficult because they are caused by complex combinations of several things going wrong. Efforts to be proactive may be sincere, but without knowledge of the causes of emergencies, they often fail to prevent further problems.
Tools and Techniques, Fundamentals, High Availability
2:20pm to 3:10pm
Room 209
  • New and Trending Topics
Peter Zaitsev (Percona)
The tremendous data growth of the last decade shows no signs of stopping. In order to effectively handle this trend, database technologies can implement a number of approaches. Enabling data compression is by far the most common and important one. While compression in open source databases is complicated, there are many different methods, each with their own implications. In this talk we will perform a review of compression in some of the most popular open source database engines, including InnoDB, TokuDB, MongoDB WiredTiger, RocksDB and PostgreSQL.
InnoDB, Other
2:20pm to 3:10pm
Room 210
  • Operations and Management
Mark Leith (Oracle)
MySQL 5.7 now includes the sys schema by default, which builds upon the awesome instrumentation framework laid by Performance Schema. Performance Schema has had 23 worklogs completed in 5.7 alone, such as memory instrumentation, tying in transactions and stored programs in to the current statement/stage/wait instruments and wait graph, prepared statement instruments, metadata lock information, improved session status and variable reporting, the new structured replication tables, and more. The sys schema builds upon this strong foundation with easy reporting views and functions, as well a
MySQL, Metrics, Monitoring
3:10pm to 3:50pm
None

3:50pm to 4:40pm

3:50pm to 4:40pm
Ballroom A
  • Operations and Management
Shlomi Noach (GitHub)
The nature of MySQL replication implies various crash scenarios of varying availability impact. Orchestrator is an open source project that discovers, manages and recovers your MySQL replication. Attend this talk to learn how Orchestrator detects failures with minimal false positives/negatives, and figures out the best method of recovery even in complex topologies. We will discuss: - Complex topology types and crash scenarios - Common crash detection methods - Configuration based vs.
MySQL, High Availability, Replication
3:50pm to 4:40pm
Ballroom B
  • Development
Alexander Rubin (Percona), Michael Benshoof (Percona)
The goal of the talk is to review the GIS improvements in MySQL 5.7, provide some general review of the functionality, as well as some working demos.
MySQL, MongoDB
3:50pm to 4:40pm
Ballroom C
  • Operations and Management
Mark Leith (Oracle)
The MySQL sys schema was integrated fully into MySQL Server from version 5.7.7. Whether you are a DBA trying to determine where the resources are being used on your database instance and by whom, or a developer trying to figure out why your MySQL statements are running too slowly, the MySQL sys schema can help. Join this session to learn how to better use the MySQL sys schema to answer your day-to-day questions—from the original developer of the MySQL sys schema.
MySQL, Metrics, Monitoring
3:50pm to 4:40pm
Ballroom D
  • Architecture/Design
  • Scalability/Performance
Raghavendra Prabhu (Yelp)
This talk is about orchestrating Percona XtraDB Cluster (PXC) nodes atop Google Container Engine (GCE) with Kubernetes. PXC provides for synchronous replication among MySQL nodes through the WSREP (writeset replication) API, and Galera plugin which implements it, to provide group communication and configuration through extended virtual synchrony (EVS).
MySQL, Docker, Other
3:50pm to 4:40pm
Ballroom E
  • Operations and Management
Josh Varner (SurveyMonkey), David Bravant (SurveyMonkey), Mark Filipi (SurveyMonkey)
In this session we will review how the open-source database team at SurveyMonkey uses Ansible as both a configuration management and automation tool for our MySQL fleet.
MySQL, Ansible, Tools and Techniques
3:50pm to 4:40pm
Ballroom F
  • New and Trending Topics
  • Scalability/Performance
Sergej Jurecko (Biokoda d.o.o.)
ActorDB is a distributed SQL database that speaks the MySQL client/server protocol. It has a buzzword rich set of features like: - Strong consistency (not eventual consistency) - No single point of failure - ACID - Replicated using Raft distributed consensus - Uses SQLite on top of LMDB as a storage engine - Scalable to a large number of nodes What's the catch? Hint is in the name and storage engine.
Other, Sharding, Replication
3:50pm to 4:40pm
Ballroom G
  • Security
Derek Downey (Pythian), Armon Dadgar (HashiCorp)
Vault is an up and coming project by HashiCorp, the folks behind projects such as Vagrant and Consul. The goal of Vault is to decouple the handling of secrets from applications to enforce access control, encryption standards, and create an audit trail. There are several components to Vault: - Authentication such as LDAP, GitHub or custom app-id - Authorization using path-based ACL policies - Encrypted storage backend using one of several options such as Consul, etcd, Zookeeper, S3, or MySQL - Secret backends that define how secrets are stored or generated.
MySQL, Devops, Tools and Techniques
3:50pm to 4:40pm
Ballroom H
  • Sponsored
Colin Charles (Percona)
The new MariaDB Server 10.1 release has been around for 6 months. It has new unique features (at rest encryption of the database, integrated Galera Cluster, GIS enhancements), performance enhancements (optimistic parallel replication, statement timeouts, dump thread enhancements in replication), as well as better MySQL compatibility (temporal literals like 5.6). This talk will go over everything new that MariaDB 10.1 has to offer.
Other
3:50pm to 4:40pm
Ballroom H
  • Architecture/Design
  • Development
To Be Defined (-)
To Be Announced
AWS, Ansible, Tools and Techniques
3:50pm to 4:40pm
Room 203
  • Big Data
Reddy Chagam (Intel), Tushar Gohad (Intel)
PCIe SSDs are becoming increasingly popular for deploying latency sensitive workloads such as database and big data in enterprise and service provider environments. Customers are exploring low latency workloads on Ceph using PCIe SSDs to meet their performance needs. In this session, we will look at a high IOPS, low latency workload deployment on Ceph, performance analysis on all PCIe configurations, best practices and recommendations.
Tools and Techniques, Data in the Cloud
3:50pm to 4:40pm
Room 204
  • MySQL 101
  • Scalability/Performance
Yves Trudeau (Percona), Matt Yonkovit (Percona)
Every day hundreds of millions of dollars are wasted by allowing improperly tuned or misconfigured systems, misunderstood infrastructure, and inefficient IT operations to live and thrive in data centers around the globe. There are both direct and indirect costs associated with allowing these unhealthy systems to continue to exist. Small performance improvements can lead to huge overall reductions in both direct costs and resource costs. During this session, we will show you how small gains in performance impact the cost of your infrastructure.
MySQL, Optimization, Fundamentals
3:50pm to 4:40pm
Room 209
  • Architecture/Design
Vladimir Fedorkov (Astellar), Peter Farkas (Altinity)
Do we always need a full blown database for your project? When would be better to use embedded database? Would it be easy to move from SQLite to MySQL later on? How could you compare performance between MySQL and SQLite? You will get a sense of these technologies in this talk. Both MySQL and SQLite are from relational database world but they are different in many ways. How to choose the right technology for your project?
MySQL, Other
3:50pm to 4:40pm
Room 210
  • Architecture/Design
Jay Janssen (Percona)
A reference architecture shows a typical, common, best-practice deployment of a system with all the surrounding infrastructure. In the case of database clusters, this can include hosting platform, load balancing, monitoring, backups, etc. A commonly referred to Percona XtraDB Cluster reference architecture was published on the Percona blog in 2012 and is included in the current manual, but is out of date.
MySQL, AWS, High Availability

4:50pm to 5:15pm

4:50pm to 5:15pm
Ballroom A
  • Operations and Management
  • Scalability/Performance
Luis Soares (Oracle)
MySQL Group Replication is a MySQL plugin that brings together group communication techniques and database replication, providing both high availability and a multi-master update everywhere replication solution. At its core is a set of group communication primitives that act as the building block to create reliable, consistent and dependable messaging between the servers in the group. This allows the set of MySQL servers to coordinate themselves and act as a consistent replicated state machine.
MySQL, High Availability, Replication
4:50pm to 5:15pm
Ballroom B
  • Architecture/Design
  • Scalability/Performance
David Murphy (Percona)
MongoDB’s basic unit of storage is a document. Documents can be of any form and change on the fly, leading to the term "schema-free" data structures. This means that we have several viable alternatives to the normalized, relational model. In this talk, we’ll discuss the trade-off of various data modeling strategies in MongoDB. You will learn: - How to work with documents - How to evolve your schema - Common schema design patterns - How to avoid common anti-patterns before they burn you - Deal with schema versions like a pro
MongoDB, Optimization, NoSQL
4:50pm to 5:15pm
Ballroom C
  • Architecture/Design
  • Development
Sveta Smirnova (Percona)
Although EXPLAIN FORMAT=JSON was first presented a long time ago, there still aren't many resources that explain how and why to use it. The most advertised feature is visual EXPLAIN in MySQL Workbench, but this format can do more than create nice pictures. It prints additional information that can't be found in good old tabular EXPLAIN, and can help to solve many tricky performance issues. In this session, I will not only describe which additional information we can get with the new syntax, but also provide examples showing how to use it to diagnose production issues.
MySQL, Optimization, Tools and Techniques
4:50pm to 5:15pm
Ballroom D
  • Operations and Management
  • Scalability/Performance
Changho Choi (Samsung Semiconductor Inc), Praveen Krishnamoorthy (Samsung Semiconductor Inc)
RocksDB contains around 120 configurable parameters that can be modified to change the behavior of the key value store. A majority of these parameters affect the performance of the database, but it is very vague when it comes to what the parameter values should be. The default values don't match up to the actual potential of the system, thereby delivering low performance. Therefore we fine-tuned the configuration parameters by performing db_bench benchmarks for different workloads to obtain the maximum throughput performance.
Other, Optimization, NoSQL
4:50pm to 5:15pm
Ballroom E
  • Analytics
Nikita Shamgunov (MemSQL)
Companies today require a defined, scalable approach to real-time analytics to keep up with immediacy of today’s user expectations. Building these business intelligence systems can strain even the most advanced architectures, placing a higher burden on resources. The ability to deliver timely, more perceptive analytics, on constantly changing data sets presents itself as a frequent challenge. For many businesses, MySQL captures and processes the majority of the application information.
MySQL, Business Intelligence
4:50pm to 5:15pm
Ballroom F
  • Operations and Management
Peter Boros (Dropbox), Aleksandr Kuzminsky (Box, TwinDB)
How to take reliable MySQL backups. There are several problems to solve: 1. Full vs Incremental vs Differential backups 2. Streaming incremental copies 3. Protecting backup copies 4. Compression options 5. On-fly backup verification 6. Successful backup restore
MySQL, InnoDB, Tools and Techniques
4:50pm to 5:15pm
Ballroom G
  • Architecture/Design
  • Operations and Management
Michael Coburn (Percona)
Everybody talks about it, but who do you know that does it right? Keeping your database schema in sync with your application is a critical component of the Continuous Deployment model of pushing code multiple times a week/day/hour. How can you, the DBA, be expected to keep up?
MySQL, Devops, Tools and Techniques
4:50pm to 5:15pm
Ballroom H
  • Case Stories
  • Operations and Management
Charity Majors (Honeycomb)
Are you an accidental DBA? A software engineer, operations engineer, or startup founder who suddenly found yourself responsible for designing/scaling/not destroying a bunch of data? Hey, me too! We should form a support group or something. In this talk we’ll cover devops/DBA best practices from the earliest seed stages (survival, selecting the right storage layer) all the way up through what you should expect from a mature, self-actualized database tier.
MongoDB, Devops, Tools and Techniques
4:50pm to 5:15pm
Room 203
  • New and Trending Topics
  • Operations and Management
Peter Zaitsev (Percona), Sebastian Stadil (Scalr)
IT used to provide and manage infrastructure, from databases to hardware. What role does it have when application developers can get a VM from AWS, manage it themselves, and use DynamoDB for state? It could be that we don't need IT anymore.
Other
4:50pm to 5:15pm
Room 204
  • MySQL 101
  • Scalability/Performance
Marco "The Grinch" Tusa (Percona)
After some years MySQL with Galera had become the most common solution when synchronous replication is needed. The cloud and EC2 in particular had being one of the most used platform to deploy successful MySQL/Galera installations.
MySQL, AWS, Data in the Cloud
4:50pm to 5:15pm
Room 209
  • New and Trending Topics
  • Scalability/Performance
Spencer Kimball (Cockroach Labs)
A new class of database engines is emerging - databases that combine the horizontal scalability of popular NoSQL databases with the transactional consistency guarantees of a traditional relational database. In this talk, we'll take a deep dive into CockroachDB, a project that aims to bring the best aspects of Google's next generation database Spanner to the rest of the world via open source.
MySQL, MongoDB, Cassandra
4:50pm to 5:15pm
Room 210
  • Operations and Management
  • Scalability/Performance
Brian Cain (Dropbox)
Overview: Imagine that your MySQL server is alerting on too many open transactions and the history list is out of control. SHOW PROCESSLIST displays a long line of connections in a sleep state with NULL info. SHOW ENGINE INNODB STATUS shows active transactions, but again there is nothing to determine what is running behind that “Sleep”. What happened within the connection? Was it just an INSERT, UPDATE, or SELECT, or was it a long string of mixed statements waiting on the commit or rollback? If you kill the connection, what might be lost?
MySQL, InnoDB, Focus Areas

5:15pm to 5:40pm

5:15pm to 5:40pm
Ballroom A
  • Architecture/Design
  • New and Trending Topics
Justin Cunningham (Yelp)
Yelp moved quickly into building out a comprehensive service oriented architecture, and before long had over 100 data-owning production services. Distributing data across an organization creates a number of issues, particularly around the cost of joining disparate data sources, dramatically increasing the complexity of bulk data applications.
MySQL, Tools and Techniques, Replication
5:15pm to 5:40pm
Ballroom B
  • Operations and Management
Igor Canadi (Facebook, Inc.)
We have been running MongoDB 3.0 with RocksDB storage engine (MongoRocks) at Parse since March of last year (2015). To celebrate the one year anniversary, we plan to share some lessons learned about running MongoRocks in production. We'll share interesting war stories and talk about performance optimization. We'll also talk a little bit about RocksDB internals and what counters are most important to watch for. Audience should come away from the talk ready to get their feet wet with MongoRocks.
MongoDB
5:15pm to 5:40pm
Ballroom C
  • New and Trending Topics
  • Scalability/Performance
Stewart Smith (IBM)
This session presents the technical reasons why you might want to consider the POWER architecture for MySQL/MariaDB production workloads. MySQL on POWER8 was the first hardware/software combination to run 1 Million SQL Queries per second, but that was an artificial workload (PK lookups) and not indicative of any real world performance so this session tackles features of the POWER architecture, the POWER8 processor and OpenPower machines that can make a difference in production environments. We'll look at: - raw compute power - many cores, many threads, a couple of sockets - mem
MySQL, High Availability
5:15pm to 5:40pm
Ballroom D
  • Security
MacLane Wilkison (ZeroDB)
Encrypting data on the client side before it reaches the server appears to be very useful for protection against data breaches and meeting regulatory requirements. However, it is challenging to implement it for databases without losing functionality. We compare our [ZeroDB] approach to other ways of implementing it (deterministic encryption with SQL/NoSQL databases, CryptDB). We show security strengths and weaknesses and performance implications of these approaches and appropriate areas to apply them.
Python, Tools and Techniques, Data in the Cloud
5:15pm to 5:40pm
Ballroom E
  • Scalability/Performance
Rick James ((self))
What are the _useful_ parts of PARTITIONing? Not much. I will briefly present the 'only' four use cases that exist for this decade-old feature of MySQL. I will give you insight into the written -- and unwritten -- limitations and gotchas of PARTITIONing. Plus a few hints about the "native" partitioning coming in 5.7 and where it is headed. Come to learn about PARTITIONing, or to understand why it is not doing all that you expected. Or come to challenge me with a 5th use case.
MySQL, Optimization
5:15pm to 5:40pm
Ballroom F
  • New and Trending Topics
  • Scalability/Performance
Dan Lynn (AgilData Inc)
Is it time to scale out MySQL? Shard now or later? Struggling with performance issues? With the current landscape of tools, analyzing MySQL performance can be time consuming and expensive. It’s about to get much easier. And FREE! At Percona Live AgilData is launching the Gibbs MySQL Scalability Advisor. Gibbs is a new free online service that lets you measure your database performance. Gibbs will point out weaknesses and opportunities for improvement and can help you plan for growth or measure against changes in your application.
MySQL, Tools and Techniques
5:15pm to 5:40pm
Ballroom G
  • Architecture/Design
  • Big Data
Ewen Cheslack-Postava (Confluent)
MySQL's binary log contains a treasure trove of event data, recording every change made to the database. However, normal database queries operate over, and mutate, only the current state of the database. This is a convenient interface for many applications, but a large amount of useful data that could be processed, analyzed, and used to make business decisions is lost if you never extract it from the binary log. In this talk, I'll describe how to combine MySQL, Apache Kafka, and Kafka's new import/export tool, Kafka Connect, to leverage this data.
MySQL, Other, Tools and Techniques
5:15pm to 5:40pm
Ballroom H
  • Architecture/Design
  • Scalability/Performance
Giuseppe Maxia (VMware)
This is a broad comparison of the methods to deploy MySQL. It will show the pros and cons of deploying database services in * dedicated servers (bare metal); * virtual machines; * containers. For each category, attendees will see the advantages, the disadvantages, and a direct comparison with the rest of the field.
MySQL, Docker, Replication
5:15pm to 5:40pm
Room 203
  • Architecture/Design
  • Operations and Management
Paige Liu (Microsoft), Roger Levy (MariaDB )
Every day we see more and more enterprise workloads starting in the cloud. It’s a clear indication that more companies are enjoying the benefits that the cloud brings, including agility, scalability, and cost effectiveness. It’s never a trivial task to set up a database cluster that is highly available and performs well. It can be even more challenging in the cloud as you need to have the knowledge about the infrastructure the cloud provides. Microsoft Azure is one of the few hyper-scale public cloud on the market.
MySQL, Devops, Data in the Cloud
5:15pm to 5:40pm
Room 204
  • Architecture/Design
  • MySQL 101
Michael Patrick (Percona)
We will discuss various topologies for achieving High Availability with MySQL such as Percona XTRADB Cluster, DRBD, MHA, MySQL Orchestrator, and others. Each solution has advantages and challenges. Attendees will gain a deeper understanding of how to choose the best solution for their needs while avoiding some of the pitfalls of making the wrong choices. Avoid the costly mistakes that commonly cause outages and lost revenue. Plus get the latest and greatest developments on the technologies!
MySQL, High Availability, Replication
5:15pm to 5:40pm
Room 209
  • Scalability/Performance
Dave Nielsen (Redis Labs)
When your application gets hammered by users or your operational system needs a boost, there is no faster database to help you increase throughput and reduce latency than Redis. This is part of the reason Redis has become one of the fastest growing databases and the most popular database deployed in containers. In this talk, Dave Nielsen, Developer Advocate at Redis Labs, will share the functions and data structures used by Redis developers to solve common scalability challenges.
Redis, Fundamentals, NoSQL
5:15pm to 5:40pm
Room 210
  • Big Data
  • Scalability/Performance
Kentoku SHIBA (Spiral Arms., Ltd)
Spider is a Storage Engine of MySQL and MariaDB for database sharding. This Storage Engine is 7 years old and bundled in MariaDB 10.0 and later. Spider supports the following things. - Dividing a huge data into multiple servers. (sharding) - Dividing a lot of accesses into multiple servers. (sharding) - Joining tables on multiple servers. (cross shard join) - Parallel processing on multiple servers. (parallel processing) - Use bunch of databases as a database. (federating) - Design and control data redundancy per table and partition.
MySQL, High Availability, Sharding
6:00pm to 7:00pm
None
Birds of a Feather Sessions.

6:01pm to 7:00pm

6:01pm to 7:00pm
Ballroom A
  • Case Stories
  • Operations and Management
Sveta Smirnova (Percona)
MySQL bug #20786 received birthday cake for its 5th year anniversary. And this bug is still "Open"! Bug #989 required full rewrite of Runtime code and 6 years of hard work. Bug #23427 was fixed in just two days. Why some bugs get fixed quickly and others wait years? What to do if MySQL bug affects you? What do all these statuses in the bugs database mean? In this session we will discuss how to open bug report, which information to provide, so it gets noticed and verified effectively.
MySQL, Devops, Other
6:01pm to 7:00pm
Ballroom B
  • Development
  • Scalability/Performance
Michael Wang (Salesforce)
mysql_use_result() or mysql_store_result() is the call you make after mysql_query() call to indicate how you would like to retrieve the data. You may not recall that you have made the call or have indicate as such, that is because mysql_store_result() is the default. mysql_store_result() as the name suggests stores the whole result set into the client memory. This has two consequences: first of all it uses large amount of memory on the client side PROPORTIONAL to the number of rows to be retrieved, and it could be out of memory if there is no sufficient memory.
MySQL, Ruby, Programming
6:01pm to 7:00pm
Ballroom C
  • Architecture/Design
Martin Arrieta (Pythian)
Many companies usually use HAProxy as a fail-over option for Percona XtraDB Cluster, but there are some considerations that you might have in count before do it. In this BOF we will discuss the different environments and architectures where you can use it and where you should not.
MySQL, High Availability
6:01pm to 7:00pm
Ballroom D
  • Operations and Management
  • Security
David Busby (Percona)
Similar to the recent MySQL Trojan (https://www.percona.com/blog/2015/10/29/mysql-trojan-chikdos/) infections we'll take a look at the complete stack compromise. - Looking over a purposefully vulnerable web application - Information gathering - Attack persistence - Attack pivoting - Metasploit - Escalation attacks on MySQL - How "bad configuration" allows malicious UDF injection on MySQL - "popping" a shell on the MySQL system (which is only addressable from the application nodes)
MySQL, PHP, Other
6:01pm to 7:00pm
Ballroom E
  • New and Trending Topics
Dave Stokes (Oracle)
Please come give feedback on current MySQL products and those in development, including MySQL 5.8 and the recently announced document store!
MySQL
6:01pm to 7:00pm
Ballroom F
  • Case Stories
  • Operations and Management
Matthias Crauwels (Ghent University)
Surely there will be a number of DBA's present that are new to the job. We could sit down and discus the problems we face starting as a DBA. A number of subjects could be relevant: - what is your backup strategy? - do you do replication? and if so, what's your topology? ... Maybe we can learn from each other and maybe we dare to ask some questions we are too afraid to ask after a presentation...
MySQL, Tools and Techniques, Replication
7:00pm to 8:00pm
None
To Book your seat at Pedro's Community Dinner, Click here.