Percona Live 2017 Open Source Database Conference

April 24 - 27, 2017

Santa Clara, California

Slides from Percona Live

Troubleshooting performance issues can be a bit tricky, especially when you’re given a broad statement that the database is slow. Learn to direct your attention to the correct moving pieces and fix what needs your attention. Learn how all this is done at Percona, what we monitor and track, and the tools we use.
Proper indexing is key to database performance. Find out how MySQL uses indexes for query execution, then come up with an optimal index strategy. In this session you’ll also learn how to know when you need an index, and to also get rid of indexes that you don’t need to speed up queries.
Go through a schema review, including reviewing large tables, looking at indexes, and so on. This will help you profile your data and schema, which will help with capacity planning or even application design.
Today's datasets are growing at an exponential rate. Collection, storage, analysis, and reporting are becoming more challenging, and the results more valued. A decade ago, RRDTool's algorithms were well suited to our requirements, but they fall short of scaling to current demands. A new direction is needed, one that prioritizes write-optimized storage, and that scales beyond a single host. This presentation will provide an...
This presentation will discuss the features of MySQL/MariaDB that when enabled and used improve the default usage of MySQL. Many cloud-based applications fail to - Using appropriate filesystem permissions - Using TLS/SSL for connections - Using TLS/SSL with MySQL replication - Using external authentication plugins (LDAP, PAM, Kerberos) - Encrypting all your data at rest - Monitoring your database with the audit plugin -...
Honeycomb combines the raw accuracy of log aggregators, the speed of time series metrics, and the flexibility of APM (application performance metrics) to provide the world's first truly next-generation analytics service. Originally modeled off of Facebook's Scuba data platform, it has spun off to become an intuitive, delightful tool for exploring every part of your stack, from debugging slow queries to db internals, from...
You’ve deployed automation, enabled automatic master failover and tested it many times: great, you can now sleep at night without being paged by a failing server. However, when you wake up in the morning, things might not have gone the way you expect. This talk will be about such surprise. Once upon a time, a failure brought down a master. Automation kicked in and fixed things. However, a fancy failure, combined with human errors,...
CRM platform Appboy uses MongoDB for its storage, with a Redis caching layer used in a variety of ways to maximize performance for the real-time messaging and analytics we view as crucial to a communication platform. Niki Castle, a software engineer at Appboy, will discuss the way we pair these technologies to power Canvas, our user journey offering, and delve into some of the most interesting and complex constraint problems that we have...
Tarantool 1.7 features a new storage engine for solid state and rotating disks: Vinyl. Vinyl implements log structured merge trees, just like Cassandra or RocksDB. In this talk, I will begin with discussing log structured merge trees, their architecture and implementation challenges. I will quote Mark Callaghan and re-assert that read, write and space amplification define the new cost model for write optimized storage. Then I will...
Percona XtraDB Cluster 5.7 is our latest offering in the clustering space. GAed 6 months, back it has lots of new features and list continues to grow. * Cluster safe mode: To get rid of those errors due to experimental features. * Instrumentation through PFS: Get insight into what is happening inside PXC. * Enhanced security: Support for Encrypted table-space, MySQL SSL-cert compatibility. * Fully integrated with ProxySQL with easy...
Today we administer so much data. From traditional RDBMS and NoSQL data. Then there is all the data from stack traces, messages in log files, and that's before we start to consider all the noise that IoT presents to us. Join me as I show you how you can bring this all together with Elastic Stack. We'll see tools for collection, clustered data persistence and visualizations to make the data scientists envious.
At Uber we have many thousands of MySQL databases running inside of Docker containers on thousands of hosts. When deciding exactly which host a database should run on it is important that you avoid hosts running databases of the same cluster as the one you are placing, that you avoid placing all databases of a cluster on the same rack or in the same data center. All these things should be kept in mind when placing a database, at Uber we...
MyRocks is a Facebook project to integrate the RocksDB LSM storage engine into MySQL. In the past, we have covered basic functionality and performance benefits of MyRocks. In this talk, we'll cover new features of MyRocks necessary to meet our scaling requirements. We have reduced data load times for the storage engine and added better diagnostic capabilities. We have found and fixed various performance and data consistency issues in...
Edgestore is Dropbox’s distributed metadata store that is used by hundreds of products, services and features - both internal and external. With over 5 million QPS and 3 trillion objects stored, Edgestore is built on top of thousands of MySQL shards and it is one of the largest systems we have built in house at Dropbox. We share a single Edgestore deployment for all workloads, which has many benefits but at the same time challenges -...
When looking for the “right” RDBMS for your application, there are many variables you need to consider to ensure you make the right choice. Not all databases are created equal, and you are inevitably going to come across some performance benchmark statistics when evaluating your options. There are a confusing variety of published benchmarks out there: YCSB, Sysbench with a variety of different versions and transaction mixes like 95:5 or 50...
MySQL 8.0 introduced roles: a new security and administrative feature that allows DBAs to simplify user management and increases security of multi-user environments. Using roles is easy, once you have digested all the documentation. For the uninitiated, though, the first approach could be disappointing, and even give the feeling of not working at all. This quick demo will show some examples of how to define roles for several...
Upgrading MySQL databases do not come without risk. There is no guarantee that no problems will happen if you move to a new major MySQL version. Should we just upgrade and rollback immediately if problems occur? But what if these problems only happen a few days after migrating to this new version? You might have a database environment that is risk-adverse, where you really have to be sure that this new MySQL version will handle the...
Whether adding indexes, altering enums, changing datatypes or reorganizing tables - there are right and wrong ways when production is involved. In this session, we will go over common operations implemented in production, discuss their impacts, and appropriate strategies for minimizing downtime and customer impact to systems. * Fast operations including index adds and enum list changes (and gotchas!) * Online schema changes using...
Apache Kafka has become very popular in the last few years. Kafka is a distributed pub/sub server for passing data in real-time. It's fault-tolerant, scalable, and extremely fast. In this talk I will discuss Kafka's core design, how it shares core architectural features of most modern databases, and how it can speed up certain workloads by amazing amounts. I will go into detail about when to best deploy Kafka and where not to...
Data comes in different shapes. One of the these shapes is time series data. Time series is a very important abstraction since it can be used to describe multiple different processes. You can discover patterns in your website users behavior, capture sensor metrics from industrial equipment or track movement of celestial bodies using time series. The real power of this abstraction lies in providing a simple mechanism for different types of...