The Database Billion

When Red Hat announced $1 billion in annual revenue earlier this year, I was cheering for a couple of reasons. First, after having spent six great years at Red Hat championing the adoption of Linux in the enterprise, I was thrilled for my many former colleagues. But more importantly, I was cheering because the perfect storm of elements that helped Red Hat achieve this milestone is now brewing in the database market. Mark your calendars on this prediction:  The relational database market is where the next open source billion will come from.

Linux champions were able to capitalize on the perfect storm of three big shifts that happened together:

  1. Standardization around largely similar technologies (Unix & Linux) that enabled much easier migration from platform to platform, preserving precious IT and sys admin skills.
  2. Acceleration of Open Source Software code contributions as developers from inside companies started pumping into the public realm massive amounts of really great code. And in turn, these same companies started considering OSS alternatives and questioning costly, proprietary lock-in.
  3. Major platform shift driven by commoditization, fueled by Intel’s x86 innovation to replace Unix boxes and thousands of dollars of pure savings per server.
.

Voila!  Disruption… and an industry transformed.

 

This same storm is brewing now in the database sector, where enterprises are even more deeply invested in their technology and the savings and other benefits of freedom are far more significant. This is how the big three shifts are playing out here:

  1. Standards-based technology in the form of SQL is as ubiquitous as any other language on the planet. But in a successful plan to create lock-in, DBMS vendors have customized their versions of procedural languages, thus deeply connecting applications (and developers) to only one database—just like the Unix forks (Solaris, HP-UX, AIX) were exposed by a Linux alternative, databases like Oracle are now exposed by PostgreSQL, the new enterprise standard for open source database.

 

  1. Developers are on a tear making extraordinary code contributions and adding new features to PostgreSQL, which is driving massive adoption across enterprises and governments worldwide.

 

  1. The extraordinary platform shift to cloud computing is the most significant change in modern computing and along with that, new platforms and new software stacks are emerging with PostgreSQL as the centerpiece.

 

  1. There is a fourth factor at play. Big Data demands are soaring along with major new IT spending on new NoSQL and analytical software technologies. So how are companies funding this move? By replacing their big spending on old proprietary RDBMS database vendors with much smaller spending for open source alternatives like PostgreSQL.

The Prediction:  Disruption.  Bye-bye Oracle, Hello PostgreSQL

 

.The PostgreSQL database has already begun to benefit significantly from this perfect storm of circumstances. PostgreSQL, or Postgres for short, is the enterprise-class open source database used by some of the largest companies in the world to handle all types of workloads. Companies like VMware, Microsoft (through its acquisition of Skype), Apple and Facebook (through its acquisition of Instagram) are all PostgreSQL users. So too are the Federal Aviation Administration (FAA), Fujitsu, NTT, Sony-Ericsson, Sony Online Entertainment, the US State Department and Yahoo.

Postgres has had decades of hardening and development by an amazingly talented and committed community as well as a fast-growing, supportive ecosystem of database specialists. This development community has contributed increasingly sophisticated features and capabilities that enterprises require in a database, including:

  • Binary log replication (Streaming replication) – both asynchronous and synchronous for simple setup of disaster recovery and reporting systems.
  • Cascading replication for supporting complex replication topologies across the globe.
  • Foreign table support for easy integration with other databases and data sources.
  • Unlogged tables for performance gains.
  • Index only scans, which allow queries to be satisfied solely by reading an index thus significantly improving performance.
  • Increased CPU core scalability to allow Postgres to scale better as hardware advances result in more powerful machines.

Over the past year we estimate that www.postgresql.org and www.enterprisedb.com, which built a series of one-click installers to ease downloads and hosts the software for downloads, have together generated over 5,000,000 downloads of Postgres and associated components. This is an enormous increase over previous years.

Premium Price or Free?

.

The growing popularity of Postgres illuminates the widespread acceptance and expansion of this open source database as the best price-performance alternative to the expensive proprietary incumbents. Where Postgres now provides the same quality and capabilities for a fraction of the cost, it’s becoming almost impossible for IT executives to justify running all workloads in the status quo.

Just look at the math to better understand how hard it is to rationalize the higher spend for all workloads. Oracle lists for $47k per core (before core count adjustments and costly add-ons like Partitioning, Spatial, Active Data Guard, Advanced Security and Oracle Enterprise Manager). And that’s before you factor in the 22 percent annual maintenance fee.

PostgreSQL on the other hand is available from a variety of support sources for a fraction of that price. Many users paying only several thousand dollars per server for a full-featured, high performance Oracle alternative, a difference so huge that users can no longer ignore the savings opportunities.

To launch the needed changes to database spending, big users of Postgres have borrowed a page from the Linux playbook—writing new apps to run on Postgres and where possible moving existing non-mission critical apps over at renewal time. The results: quick wins, comparable performance and big savings.

This adoption is shockingly similar to patterns we experienced in the wave of transformation and disruption of Linux. This is the jet fuel that will make PostgreSQL the next open source billion. Mark your calendars.

About the Author

Ed Boyajian is president and CEO of EnterpriseDB, which provides enterprise-class PostgreSQL products and services to help IT organizations succeed with the world’s most advanced open source database.