M|18

Sessions

Join the conversation
WATCH THE ON-DEMAND VIDEO RECORDINGS

M|18, MariaDB's user conference, featured sessions that offered practical solutions for database administrators (DBAs), developers, DevOps, cloud technologists and decision-makers on topics ranging from the MariaDB kernel itself, to scale-out Big Data and container configurations, to the latest in how temporal processing raises the bar in security.

All of the session recordings are now available on demand. To view the recording, select the session you'd like to watch and click the "Watch Now" button.

Sessions
Watch Video
Watch Video

ServiceNow, faced with unprecedented scalability challenges as it became the fastest growing enterprise software company over $1 billion, required database infrastructure capable of supporting petabytes of data and billions of queries. How did they do it? Why did they switch to MariaDB?

Tim Yim

Speaker

Tim Yim
ServiceNow
Director of Operations
Watch Video

The largest bank in Southeast Asia, DBS Bank returns to the MariaDB keynote stage to share its success, migrating from proprietary products to open source with MariaDB. Ng Peng Khim, Managing Director, Head of Institutional Banking Group & Future Ready Technology and Joan Tay Kim Choo, Executive Director of Technology, will share DBS Bank’s corporate internet banking use case and how the team deployed more than 50 percent of their critical applications on MariaDB in the last two years.

Ng Peng Khim

Speaker

Ng Peng Khim
DBS Bank
Head of Institutional Banking Group & Future Ready Technology Group Technology & Operations
Watch Video

MariaDB ColumnStore’s architecture is a meld of MariaDB’s traditional row based database server with a distributed high performance analytical columnar storage engine. This session is for MariaDB users so they can get to grips with the the MariaDB ColumnStore engine and how it works. It will cover:

  • Use cases
  • Query processing
  • Bulk data insertion
  • Distributed partitions
  • Optimizing queries

We will also go over the new features in MariaDB ColumnStore 1.1 and a sneak peek at some upcoming features.

Watch Video

As businesses become more and more data intensive, the cost per transaction becomes an important metric. The combination of MariaDB and Intel® technologies is extremely powerful in this age of distributed computing. In this session, we will discuss the use of shared, log-structured storage to support the persistence requirements of databases. We will outline the details of this approach and provide some early results based on an operational MariaDB/MyRocks implementation. Furthermore, you will learn how our cooperation supports the transformation of transaction performance and cost by optimizing the combination of MariaDB running on the Intel® Xeon® processor Scalable family and Intel® Optane™ SSDs taking advantage of 3D XPoint™ memory media. We will also describe how Intel® FPGAs are accelerating on-prem cloud-scale DBaaS.

David_Cohen

Speaker

David Cohen
Intel
Storage Solutions CTO and Senior Principal Engineer, Data Center Solutions Group
Mike Strickland

Speaker

Mike Strickland
Intel
Director, Solutions Architect, Intel Programmable Solutions Group
Watch Video

This session will outline the high availability options available in MariaDB TX, highlight the pros and cons of different approaches, explain the trade-offs involved, and show how to optimize for consistency and/or performance.

Watch Video

A traditional problem of relational databases is storing similar data with different attributes. For example, querying a product catalog when different products have different properties (e.g., color for shirts, length for pants). You can store everything in the same table or create multiple tables, but both options have their issues. There are other design patterns as well. MariaDB Server, with dynamic columns and JSON functions, introduces new options. In this session, we will evaluate the pros and cons of these approaches based on different use cases.

Watch Video

This session will review new features and improvements available in the MariaDB TX platform, focusing on MariaDB Server 10.3 and MariaDB MaxScale 2.2.

Watch Video

This session will provide an introduction and overview of the MariaDB AX platform, highlight updates to the ColumnStore storage engine and provide an update the latest tools and connectors, including the new streaming data adapters.

Watch Video

The Ansell Guardian® service helps companies select the right personal protective equipment solution for improved safety, productivity and cost performance, resulting in a 65% injury reduction and savings of over $149 million for our customers.

In early 2017, we seamlessly replaced the global geo-distributed data synchronization processes supporting the Guardian application suite with MariaDB TX, resulting in better performance and vastly reduced support issues from the previous MySQL-based solution. We will discuss our preparation for this successful implementation and how MariaDB TX with clustering supports the Ansell Guardian application – and the in-house scripting tools we developed to simplify administration of it.

Greg Henderson

Speaker

Greg Henderson
Ansell Healthcare
Senior Manager, Architect - Global Business Systems
Louis Zircher

Speaker

Louis Zircher
Ansell Healthcare
Director, Ansell Guardian® Technology Solutions
Watch Video

It’s easier than ever to migrate from Oracle to MariaDB TX, with new features introducing compatibility and reducing the effort necessary to conduct a successful migration. In this session, we will discuss our experiences with customers migrating from Oracle, and the lessons learned we’ve learned. In addition, we will introduce the Oracle PL/SQL compatibility and sequence support in MariaDB Server 10.3, noting what Oracle PL/SQL syntax is supported, and show how Oracle PL/SQL stored procedures can be run on MariaDB Server.

Watch Video

The goal of this session to help attendees understand when to use MyRocks and how. MyRocks is storage engine developed by Facebook. It was engineered to meet the highest compression and write efficiency requirements. While it was alpha in MariaDB 10.2, it is expected to be GA in MariaDB Server 10.3. In this session, Sergei will cover everything from intended use cases and expected benefits to important differences when compared with InnoDB.

It will conclude with walkthrough of practical considerations when migrating and/or using MyRocks:

  • Tuning for performance
  • Migrating data from InnoDB
  • Backing up and restoring data
  • Replication considerations
  • Using write-optimized operation modes
  • MyRocks differences in MariaDB
Watch Video

With bulk data adapters for C++, Java and Python, developers can bypass the SQL interface, parser and optimizer for high-throughput, low-latency ingestion of bulk data – ideal for storing machine learning results, collecting clickstream data, importing scientific data and more. This session will provide an overview of how the bulk data adapters work and show how to use them with examples. We’ll also give a demo of how we’ve integrated this with Spark.

Watch Video

This session will focus on the creation a reusable reference architecture at Nokia to simplify the deployment of MariaDB TX in multiple data centers to maintain high availability. We’ll start with the high availability and failover requirements for geo-distributed applications. We will then examine the high availability and failover options provided out of the box before describing our reference architecture and showing how native automatic failover can be combined with real-time events and external scripts to maintain HA within and between data centers.

Rick Lane

Speaker

Rick Lane
Nokia
Consulting Member of Technical Staff
Watch Video

PolarDB uses shared storage to provide read scalability, using InnoDB redo logs for physical replication and storing physical page level operations for crash recovery. PolarDB extends this functionality, allowing multiple read replicas to be deployed for read scaling. In this talk we’ll take a deep dive into InnoDB internals and explain the changes we made to the core InnoDB code. We’ll touch upon design issues around logging, crash recovery, buffer pool management, MVCC, DDL synchronization and more.

Watch Video

This session will introduce the point-in-time querying functionality added in MariaDB Server 10.3, using practical examples to show how SQL can be used with AS OF and system versioned tables to query data as it was at a specific point in time.

Watch Video

This session will provide an overview of the Spider storage engine (GA in MariaDB Server 10.3), explain how it works, highlight recent updates and show how to configure and deploy MariaDB Server with scalable storage using distributed partitions.

Watch Video

MariaDB AX now includes two streaming data adapters. The first consumes messages from MariaDB MaxScale change-data-capture streams – data written to MariaDB TX is automatically and continuously replicated to MariaDB AX. The second consumes message from Apache Kafka topics. This session will introduce these streaming data adapters and explain how they can be deployed and configured to ensure data in MariaDB AX is always up to date and without the need for manual and/or batch processes.

Watch Video

In this session, we will explain our evaluation criteria for selecting a new database (e.g., security considerations in the financial sector) and why we chose MariaDB. We will then focus on how we migrated from Oracle and database-specific development to MariaDB and database-agnostic development. Finally, we will conclude with a summary of where we are in the migration process: implementation, performance and production.

William Wood

Speaker

William Wood
Financial Network, Inc.
Director of Database Architecture
Watch Video

An important part of hardware cost and efficiency is performance tuning and optimization. This session, for developers and DBAs, will describe how to optimize everything from the OS to the network to the database (and queries) to get the best performance possible.

Watch Video

With the rise of GPS-enabled devices and beacons, comes the need, opportunity and use cases to extend applications with location-based services. This session will introduce the latest GIS features in MariaDB Server and explain the supported GIS data types and how to index/query them.

Watch Video

This session will provide attendees with common use cases for clustering, a comprehensive explanation of how clustering works with Galera and introduce new features and improvements available in Galera Cluster 4.

Watch Video

Hear directly from MariaDB AX users from diverse industries. Our group of panelist are building innovative use cases such as genome analysis for cows and pigs, healthcare program cost-effectiveness analysis to improve public health and customer behavior analysis to provide high-quality customer support. In this session, you will learn why these companies chose and how they implemented MariaDB AX.

Jack Sprague

Speaker

Jack Sprague
Center for Information Management
Enterprise Architect
Watch Video

This session will feature a panel discussion and open Q&A on successfully adopting OSS in the enterprise. Red Hat, Copart and Financial Network, Inc. will share their experience with OSS: why they chose to adopt it, how important it is, how they got started, the business and technical challenges they faced and how other enterprises can succeed with OSS too.

Watch Video

PolarDB is a next-generation, cloud-native storage engine for Alibaba Cloud. PolarDB for MyRocks extends PolarDB to support MyRocks on shared storage, using RocksDB logs for replication. In this session, we’ll explain how we solved many problems for deploying MyRocks on shared storage: log replication, converting system tables, cache and DDL replication, MVCC and the new RocksDB log format.

Watch Video

It’s one thing to run a database on a single containers, it’s another to run a replicated/clustered database on multiple containers. This session will provide an overview of the requirements for running databases on containers and show how to configure and deploy a MariaDB TX cluster on Docker.

Watch Video

In this session, Sergei Petrunia and Igor Babaev will provide a technical overview of query optimizations for views, derived tables and common table expressions – embedded queries, late materialization, condition pushdown, lateral and mergeable derived tables and more – while Vicentiu Ciorbaru will covers query optimizations for window functions.

Watch Video

With scalable, high-performance analytics readily available, organizations of all kinds are moving beyond reporting, building decision support systems and using predictive analytics to identify outcomes and provide recommendations. In this session, we will explore innovate use cases in the finance, healthcare, telecommunications and manufacturing seconds and highlight the most interesting case studies.

Watch Video

Christian Lees, CISO, will provide an illuminating, “behind the curtain” look at threat actor attribution and the underground economy, information gained from years of research into the dark web. He will examine methods and provide examples of how InfoArmor harvested data, specifically compromised credentials, from the underground economy. Steve Olson, lead engineer, will then expand on the technical details of how data is ingested from the dark web.

What we learned from growing by a billion rows:

  • Query and table optimization
  • Application optimization
  • Monitoring bad queries via application logging and Montog
  • Python pyramid -> sql alchemy ORM -> Zope transaction manager
  • MariaDB consulting role -> advice, config, admin work
  • Replication, read/write splits -> MariaDB MaxScale
  • Hardware configuration needed
Watch Video

In this session, we will explain how we implement database security at Tencent Cloud, focusing on encryption and auditing with technical explanations of the security architecture, the plugin implementations and how we use audit logs to diagnose database issues.

Watch Video

The first half of this session will provide an overview of common table expression (CTEs), the SQL syntax, use cases for non-recursive CTEs and query optimizations for CTEs, including condition pushdown. The second half of this session will focus on recursive CTEs, use cases such as transitive closures and an explanation for how loops are handled and stopping recursion.

Watch Video

This session will introduce the automatic failover feature being developed in MariaDB MaxScale, providing an overview of failover concepts, a step-by-step explanation of the failover process and how to configure it.

Watch Video

MariaDB AX leverages distributed, columnar storage for performance and scale, but what if you want to get started without having to run MariaDB AX in a scale out environment? In this session, we will walk you through the process of building a multi-node MariaDB AX cluster on a single machine using Kubernetes and containers. With a straightforward and practical introduction to both MariaDB AX and Kubernetes, we will give attendees the know-how to build and test a real-world deployment right from their laptop.

Watch Video

Due to exponential growth in its international business unit, Copart face significantly higher database costs if we continued to use IBM DB2. We will explain why we chose to standardize on MariaDB TX, how we convinced operational and development teams to adopt it, the challenges we faced and how we are now evaluating MariaDB AX as a replacement for our current analytics database too.

Watch Video

Facebook created MyRocks, and it is now available in MariaDB. MyRocks is a space and write optimized storage engine built on top of RocksDB. The initial objective was to replace InnoDB with MyRocks in the users database (UDB), reducing the amount of storage space needed by half. It was a success, and now Facebook is trying to make it better and help increase adoption in other services. In this session, Yoshinori will explain how Facebook replaced InnoDB with MyRocks in UDB, and highlight the features being developed and planned for future releases of MyRocks.

Watch Video

In this session, we will explain how when, where and how to use windows functions (e.g., to optimize queries with self-joins and related rows) as well as how to create your own user-defined functions and user-defined aggregate functions. We will cover everything from implementation details and performance advantages to use cases and practical examples.

Watch Video

Johan Wikman, the lead developer for MariaDB MaxScale, will describe the architecture behind MariaDB MaxScale and explain how it has evolved with each release, improving performance and scalability.

Watch Video

Creating a reporting and analytics solution to manage over 200 retailers and 60 million loyalty cards is a complex problem, especially when you have to integrate multiple components from different vendors – and with a consistent architecture. In this session, we’ll look at the real life use case for a large, UK-based customer loyalty program running reporting and analytics as a service on Microsoft Azure. The solution is built on MariaDB AX, Analysis Services and other products and services to delivery an intuitive front end.

Watch Video

When it comes to building scalable infrastructure, making resources expandable is always the right choice. In this session, we will discuss our shift to containers and high availability. We will provide an overview of our infrastructure and environment, introduce our service discovery solution and reveal what we call “Backend High Availability Pillars” with MariaDB Galera as example.

Watch Video

The best way to maintain high availability for a database is to abstract away the underlying infrastructure. MariaDB MaxScale, an advanced database proxy with a networking core and modular architecture, enables applications to connect and query multiple database clusters as if they were a single database. This requires MariaDB MaxScale to detect changes to the database topology and state of the databases. In this session, we will explain how MariaDB MaxScale creates this abstraction, and how it is able to route database queries to the right database servers.

Watch Video

We will review different options for performing online schema changes without downtime, whether it is via built-in database capabilities or external tools like gh-ost and pt-online-schema-change, highlighting their strengths and weakness and using practical examples to provide recommendations on what to use and when.

Watch Video

This session will explain in technical detail every step in the write path for InnoDB transaction, from the very beginning to the very end. It will introduce advanced DBAs to mini-transactions and how they used to modify buffer pools, redo/undo logs, clustered and secondary indexes, and checkpoints – all the way to fsync and commit.

Watch Video

In this session, we will explain how GPU processing can accelerate analytical workloads using advanced data structures and recursion. We will then demo a preview of initial GPU support in MariaDB Server, contributed by Brytlyt, to show how response times can be cut by minutes when querying millions to billions of rows.

Watch Video

This session will explain how DBAs can use automation to simplify database administration – everything from backups and maintenance to monitoring and alerts across multiple databases and servers. We will highlight the pros and cons, discuss how we use it and introduce the tools we use. In some cases, it has made our easier. In others, not so much. We’ll be sharing best practices and recommendations based on the lessons we’ve learned.

Watch Video

The database proxy is a key component of modern database infrastructure, responsible for routing queries to the right database server, regardless of the topology. However, MariaDB MaxScale, as an advanced database proxy, is capable of a lot more – everything from caching to data masking to change-data-capture. In this session, we will walk through all of the features in MariaDB MaxScale and explain how it improves database performance, scalability, availability and security. In addition, we’ll show how MariaDB MaxScale can be extended by creating customer modules. In this case, a simple module for generating a whitelist for the firewall by inspecting SQL queries.

Watch Video

In this session, we will discuss writing stored procedures in MariaDB Server and show working code examples with a live demo using Vagrant.

Watch Video

In this session, we’ll explain in technical detail every step that occurs when data is replicated – what’s replicated, when and where. We will introduce concepts like group commit and GTIDs, the binlog format and data structure (and relay log), parallel replication threads and how the replication process fits within a transactions – what is written to disk and when, and when is the data durable.

2018 Sponsors