Rising Interest in Open Source Relational Databases
- 1. INSIGHTS Presentation Series The Rising Interest in Open Source Relational Databases Chris Foot, VP Technologies and Strategies Date: 6/22/2017Replacing traditional DBMS licensed products with open source offerings Video Presentation Inside!
- 2. The Largest Pure Play Provider of Managed Data Infrastructure Services Database Platforms SQL Server Oracle PostgreSQL* DB2 MongoDB* MySQL* Operating Systems Unix/LinuxWindows Edge Technologies SQL Server BI Oracle EBS SharePoint Exchange Environment 450+ Customers 10,000 Servers 200+ DBAs Fortune 100s Startups All Verticals Cloud Systems Amazon AWS/RDS Oracle Cloud DB DBPaaS Msoft Azure IaaS (dozens) Hybrid Cloud * All distributions We support hundreds of commercial and open source DBs! And understand benefits and weaknesses of both platforms
- 3. Lots of Open Source Alternatives Public Cloud DBaaS CLOUD NoSQL MySQL PostgreSQL OPEN
- 4. Wide Range of Pricing and Support Services • Pure Open Source o Most commonly known as Community Editions o No licensing fees o Support provided by community • Open Source With Professional Support Services o No licensing fees o Vendors provide 24x7x365 support for open source offerings • Open Source Commercial Versions o Vendors start with base code and add additional features o Link to community editions and push enterprise offerings o Customers purchasing commercial versions take advantage of additional features and can choose from 24x7x365 support options o Quickly adopt community edition new releases o Commercial vendors may/may not contribute their advanced features back to community (mostly not) • Pay as You Go o Database Platform as a Service or DBPaaS o All vendors are jumping into the cloud o Cloud providers rent their versions of open source offerings X Open Source Doesn’t Mean Everything is Free
- 5. Why Aren’t We Discussing NoSQL? March’s RDX Insights Series presentation focused on NoSQL YouTube presentation: https://youtu.be/JbyEglIvLjo Slideshare presentation: https://www.slideshare.net/ChristopherFoot/nosql-architecture-overview
- 6. DB-Engines Popularity Trends http://www.db-engines.com Open Source Popularity Increasing Commercial Decreasing
- 7. DB-Engines Top 5 http://www.db-engines.com Big drop in popularity score between #2 and #3 MySQL score exceeds SQL Server Big drop beween #1 and #2
- 8. DB-Engines Growth http://www.db-engines.com Oracle and MySQL dropping in popularity PostgreSQL and SQL Server increasing in popularity PostgreSQL overtakes MongoDB
- 9. DB-Engines Popularity By Storage Model http://www.db-engines.com Relational 80% NoSQL 20%
- 10. DB-Engines Storage Models (2013 -2017) http://www.db-engines.com Popularity leadership changed many times for NoSQL storage models Relational stays constant at 80%
- 11. NoSQL vendors desire to increase market share will drive them to compete directly with relational product manufacturers Vendors will add RDBMS-like functionality that allows their product to be more widely adopted. Those that don’t will quickly lose market share to those that do The larger relational vendors will attempt to co-opt any NoSQL technology that challenges their dominant role in the industry As they identify offerings as tangible threats, their strategy will be to ensure that the technologies used by those vendors become a component of, not a replacement for, their traditional database products Relational DBMS NoSQL DBMS General Purpose DBMS The Multi-Model DBMS Engine
- 12. Dice Insights JOB TITLES IN MORE DEMAND
- 13. • Runs on Windows and Linux • ACID compliant, object-relational • Robust procedural language – PL/pgSQL • Supports all major programming languages • Triggers, foreign keys, functions, triggers, views, materialized views, JSON, key-value, recursive • Statistics based optimizer, parallel query, secondary indexes, index-only, range/list partitioning, full text search • Log shipping, standby, master/slave, multi-master through open source and commercial add-ons • Clustering available through open source and commercial offerings add-ons • Roles, row-level security, security, labels, SSL, connection limits, LDAP, Radius • Point-in-time recovery • Lots of open source GUI administration tools available • Rapidly increasing in popularity • Robust open source development community • Many commercial support options available • Number of cloud DBPaaS offerings growing quickly o Amazon Aurora and Google SQL • Growing popularity is: o Increasing vendor competition o Growing number of vendors offering tools, integrating their apps with PostgreSQL, offering add- on features and support services PostgreSQL
- 14. • Robust set of services, administrative tools and add-on features • Provides Oracle, SQL Server conversion tools • EDB Postgres Ark allows users to seamlessly deploy on public/private clouds • EDB Postgres Enterprise Manager is monitoring product • EDB Replication Server provides multi-master • EDB Failover Manager provides automatic failover • EDB Postgres Advanced server provides native database compatibility with Oracle (PL/SQL, built-in packages, tools) • EDB Public Cloud is a DBPaaS deployment for Amazon AWS and Google Compute Platform PostgreSQL Vendor Offerings • Increasingly popular cloud development and runtime platform • Robust architecture simplifies and accelerates the development and deployment of cloud based apps • Robust, scalable, runtime environment with strong monitoring capabilities • Built on specialized Linux containers called Dynos which offer easy scalability • Dozens of utility add-ons (databases, logging, email, analytics) • Offers PostgreSQL as a Service (DBPaaS)
- 15. PostgreSQL Vendor Offerings • PostgreSQL as a Service (DBPaaS) • Deploy to Amazon AWS, Microsoft Azure, Google Compute Engine, IBM Softlayer • Automated backups, provisioning, replication configuration, monitoring tools • Support for multi-data center and multi-region • Plans based on data stored, memory, CPU, concurrent connections • Amazon RDS support for Oracle MySQL, MariaDB • Announced Aurora’s PostgreSQL compatibility in April 2017 • States that “all of the code, applications, drivers and tools you use today with your PostgreSQL databases can be used with Amazon Aurora with no change.” • Google’s Cloud SQL PostgreSQL compatibility is currently in beta • Scales to 32 processors and 200 GB of RAM • Still being improved and updated as of the date of this presentation • Original codebase was PostgreSQL • HIGHLY modified for massively parallel data warehouses • Continues as open source
- 16. • Perennial open source favorite and powerhouse • Runs on 20 platforms • Oracle “sponsors” and offers commercial editions (Classic, Embedded, Standard, Enterprise and Cluster) • Community releases minor releases every 2 months • ACID compliant • Multiple storage engines tailor DB to workloads • Wide range of GUI admin tools and utilities available • Supports all major programming languages, JSON • Triggers, foreign keys, stored procedures, functions, views • Statistics based optimizer, secondary indexes, index-only (covering index), range/list/hash partitioning, full-text search • Log shipping, standby, master/slave and multi- master • Clustering available through open source and commercial offerings • Galera cluster is known to be an excellent open source clustering solution with support services available • Roles, SSL, encryption, security plug-ins • Point-in-time recovery MySQL
- 17. • Oracle offers following commercial editions (Classic, Embedded, Standard, Enterprise and Cluster) • Standard Edition does not have advanced features (compared to more expensive versions) but does provide 24x7 support, maintenance releases • Enterprise Edition offers cloud, Enterprise Dashboard, Query Analyzer, Replication Monitor, partitioning, MySQL Router, hot backups for InnoDB, full/incremental/partial backups, full/partial/hot selective restore, PIT recovery, Enterprise Encryption, Enterprise Audit, Enterprise Firewall, Group Replication, InnoDB Cluster, Oracle OEM Admin Tool for MySQL • Robust MySQL DBPaaS cloud offering • Cluster CGE Edition provides support for high availability, heavy workloads and massive scaling • Oracle touts 99.999% availability, 200 million reads/second, 2.45 million SQL statements/second • Provides same features as Enterprise • Also offers in-memory, SQL and NoSQL APIs, multi-site clusters with active/active geo replication, auto sharding with full join, referential integrity and ACID support, partitioning, NDB Storage Engine, MySQL Cluster Manager MySQL Vendor Offerings
- 18. • Offers Percona servers for MySQL and MongoDB • Describes product as “a free, fully compatible, enhanced, open source drop-in replacement for MySQL that provides superior performance, scalability and instrumentation. “ • Provides 24x7 support for MySQL, Percona and MariaDB • Percona server includes utilities that add additional administrative capabilities and very strong visibility into MySQL performance • Historic flagship product is open source XtraDB Cluster • Percona XtraDB Backup is open source hot backup utility • Percona TokuDB is a highly optimized storage engine for Percona Server providing HA, compression and performance features • Sole focus is MySQL cloud, offering MySQL for Amazon AWS, Google Cloud Platform, IBM Softlayer, Microsoft Azure • Editions include Community, Developer, Standard and Enterprise • Leverages inherent cloud features (regions, geo data distribution) • Also provides automatic backup scheduling, automatic failover, read replicas, multi-cloud and cloud/on-premises clustering, master/slave and master/master replication, backup encryption • Advertises benefits of no cloud vendor lock-in • 100% up-time guarantee that 1 master node will be online MySQL Vendor Offerings
- 19. MySQL Vendor Offerings • Amazon RDS support for Oracle MySQL, MariaDB • States that Aurora is “fully compatible with MySQL 5.6 using the InnoDB storage engine” • Offers utilities that allow customers to convert competing MySQL products to Aurora • Also offers on-premises to cloud replication tool to facilitate cloud conversions • Provides “push button” scalability • Leverages Amazon architecture for storage, backups, geo data redundancy, read replicas, database snapshots, • PIT recovery, encryption, monitoring tools, event alerts • Need to be cognizant of feature mismatch between Aurora and traditional MySQL offerings • Provides traditional cloud benefits including automatic backups, failover • Leverages Google architecture for geo data replication • Scalability to 10TB of storage capacity, 25,000 IOPS, and 208GB of RAM per instance • Provides replication, encryption for DB and backups • Focuses on enterprise class cloud architectures • Provides MySQL as a DBPaaS fully managed service offering • Database replication and enterprise monitoring tool is an extra charge
- 20. • Created by the original developers of MySQL • Goal was to continue MySQL as a pure open source community project (reaction to Oracle takeover of MySQL) • Goal is for MariaDB to be a seamless drop in for MySQL • The product is a fork of MySQL 5.5. There will be variations between MySQL and MariaDB. Goal of the community is to keep these to a minimum • Vibrant development community that is rapidly providing additional features to the core product • Offers more storage engines than commercial counterpart • MySQL vs MariaDB feature comparisons can be found here: • Increasing number of vendors offering 24x7 support • Increasing number of vendors providing MariaDB DBPaaS • Amazon offers MariaDB RDS service • MariaDB 10.1 download includes Galera row-level master/slave and master/master clustering add on MariaDB https://mariadb.com/kb/en/mariadb/mariadb-vs-mysql-features/
- 21. Who Will Win the Database Wars? Open Source DBMS Commercial DBMS
- 22. Strengths Weaknesses • Lower up-front licensing costs • Lower maintenance costs • Growing feature set and increasing functionality • Vendors stepping in to provide 24/7 product support and maintenance • Vendor support quality can range the spectrum • Access to DBA and developer skill sets • Challenge to compete with huge, big budget vendors constantly offering new features – especially for high availability, scalability, data warehouses • Third-party software provider adoption (applications, tools and utilities) • Conversion costs • Interaction with other DB systems Open Source DBMS
- 23. Strengths Weaknesses • Strong vendor support • Robust features and functionality • Most vendors have strong cloud strategies • Strong third-party software provider adoption (applications, tools and utilities) • Access to DBA and developer skill sets • Higher up-front licensing costs • Higher maintenance costs • Vendors' licensing practices often become predatory • Complex to administer Commercial DBMS
- 24. RDX Recommends Add an open source database to your stable of products Create an open source/commercial database strategy Thoroughly understand and evaluate competing offerings Consider cloud based and on-premises alternatives Select the appropriate database(s) for initial conversions Production applications require 24x7 product support
- 25. RDX Recommends These DBS will support mission critical apps (they have for years), but go with a proven software support provider (Oracle, Amazon, EDB, Percona, CenturyLink, ClearDB) They are cost effective, robust, reliable alternatives to commercial products Realize that adding a new product will incur startup and conversion costs Oracle customers should consider EDB as a way to reduce overall DB TCO without extensive training and retooling
- 26. Policies and Procedures Monitoring and Event Management Impact on Existing Tools and Technologies Training and Education New Staffing Roles and Responsibilities Change Management Administration New DBs Will Change the Way Your Organization Provides Support Costing Models
- 27. DBMS Product Features Don’t Always Match 100% App Code Transportability Database Features Cloud DBMS vs On-Premises Distributions and Forks Vendor Feature Add Ons
- 28. The More You Have to Tailor Your DB/Application to a Distribution The harder it will become to switch vendors
- 29. Upcoming Presentations – Cloud’s Hidden Impact on IT Support Organizations, Benefits of PCI DSS Compliance The RDX Report Microsoft BI Overview and Power BI Demo, SQL Server 2017 New Features, Amazon AWS Data Migration Services and Google Cloud SQL Support for PostgreSQL LinkedIn Selecting Cloud DBMS, NoSQL Architectures, Database Security Series, Improving Customer Service 20YEARS OF SERVICE DELIVERY EXPERIENCE [email protected] Video recording of this presentation can be found on RDX’s YouTube Channel: RDX YouTube Channel Presentation