Percona Live is a jam-packed few days. There are over 100 sessions, covering a range of open source topics and featuring several tactical, hands-on demos. Obviously, you can’t attend every one of them, and parsing through the agenda can take some time.

That’s why we’ve compiled a list of what we’re calling “hot topic” sessions. These are the ones that are top-of-mind in the open source community and feature heavy-hitting speakers from Meta/Facebook, Venmo, Amazon, and, yes, Percona. They’re also the only sessions that are available for live streaming. 

Below, you’ll find the full list of “hot topic” sessions. And if you haven’t yet registered for Percona Live, the largest open source database conference in the world, head on over to our registration page. Just do it quickly – Percona Live is happening May 16th – 18th in Austin, TX. 

Tuesday, May 17, 2022

The Evolution of a MySQL Database: From a Single Instance to HA with Disaster Recovery, 10:30 AM

This session discusses how to make the transition from a single MySQL instance to multi-site high availability, as well as which solutions are best suited to changing business requirements (RPO, RTO).  

MySQL Performance Diagnostic Using Percona PMM, 11:30 AM

In this session, we show how we use Percona Monitoring and Management (PMM) at our Support department to systematically find the root cause of performance issues.  We will showcase our bottom-up approach, based on the U.S.E. method by Brendan Gregg, and will diagnose a live workload that the attendees themselves will have a chance to tweak during the presentation.

MariaDB Scaling – From Single Instance to Multiple Clusters, 1:30 PM

In this session, attendees will follow the story of the growing application with respect to database architecture – from a single-instance database on a developer’s machine to cross-DC multi-cluster HA setup using Galera, ProxySQL, and anycast routing.

Scaling Venmo Applications for Growth With Zero Database Downtime: Venmo (Paypal Inc), 2:30 PM

This talk will describe how Venmo used proxy solutions for MongoDB and Aurora-MySQL to scale with traffic, various alternatives explored and driving factors narrowing down to a chosen approach. Venmo will share how they migrated the traffic via proxy with zero downtime, the challenges faced and the solutions applied to achieve their goal. 

Looking Ahead at PostgreSQL 15, 4:00 PM 

This talk explores many of the new features that will be released in the next major version of PostgreSQL and explains how they can impact your workload. 

Data Consistency at Scale in Meta,  5:00 PM

Meta/Facebook has gone through multiple large-scale MySQL rollouts like MyRocks and 8.0 migrations. To avoid the potential data inconsistency introduced by those migrations and various other small rollouts like minor version upgrades, they implemented several large-scale consistency checking tools to help surface issues at an early stage. In this talk, they will cover how these tools work under the hood and share some stories about how they have helped in large-scale rollouts.

Wednesday, May 18, 2022

Deploying Highly Available, Durable Amazon RDS for PostgreSQL and for MySQL Databases with Multi-AZ with Two Readable Standbys, 9:30 AM

Amazon Relational Database Service (Amazon RDS) for PostgreSQL and for MySQL now support a new Amazon RDS Multi-AZ deployment option with one primary and two readable standby database (DB) instances across three Availability Zones (AZs). With this new option, Multi-AZ users can get up to 2x faster transaction commit latency compared to one with standby, typically under 35-second failovers, and additional read capacity. In this session, learn about the new option, and how it compares to other high availability options, and see a demo on how to get started.

Postgres and the Artificial Intelligence, 10:50 AM

Artificial intelligence, machine learning, and deep learning are intertwined capabilities that attempt to solve problems that defy traditional computational solutions — problems including fraud detection, voice recognition, and search result recommendations. While they elude simple computation, they are computationally expensive, involving the calculation of perhaps millions of probabilities and weights. These computations can be done outside the database, but there are specific advantages of doing machine learning inside it, close to where the data is stored. This presentation explains how to do machine learning inside the Postgres database.

Multi-Tenant Kubernetes Cluster With Percona Operators, 11:50 AM

Percona Operators provides a great way to run databases on Kubernetes. But what if you want to run multiple databases on a single k8s cluster? In this talk, we will explore how this is possible and also walk through some of the best practices that make multi-tenancy of DB on k8s a smooth experience. 

Efficient MySQL Performance, 2:00 PM

MySQL performance can be challenging for new software engineers because where does one begin? Even experienced engineers can find MySQL performance challenging because it’s not their area of expertise. This session covers the path to learning and achieving better MySQL performance by focusing on the most important topics for software engineers using MySQL, not aspiring DBAs. 

Database Resiliency,  3:00 PM

This session covers cutting-edge strategies in handling scalability, resiliency, and high availability, building fallout tolerant database solutions, preventing global outages by preventing the propagation of failures, scaling a high-volume, high-transaction based database solution,  and safely performing complex database changes in distributed database solutions. 

Register for Percona Live