A Blog by Jonathan Low


Jan 8, 2015

How Open Source Databases Are Gaining - and Chipping Away at Oracle's Dominance

Oracle so dominated databases for so long that challenges to its hegemony once seemed too far-fetched to even contemplate. 

But as so often happens life and especially in tech, things changed. There are, of course, technological and strategic developments that undermined the preeminence of the big database companies. Certainly the emergence of Amazon and Google, the role of the cloud and the increasing growth of mobile all contributed.

The more fundamental evolutionary metamorphosis may have been in the emerging role of data as a driver of performance rather than merely a supplemental support. In order for that transformation to gain traction, the sources and economics of data access and transfer had to change. This in order to facilitate the growth in data available, the speed with which it has to be accessed and the purposes to which it is being applied.

This will, in the long run, create more opportunity for data usage, storage and profit, as well as the growth of ancillary services focused on interpretation and presentation. As in so many other facets of tech's evolution, the optimal path to profitability and predominance may not take the traditional course. JL

Klint Finley reports in Wired:

The three fastest growing databases of 2014 were all open source.
The three fastest growing databases of 2014 were all open source, according to a new report from DB-Engines, a site that tracks popularity in the rapidly changing database marketplace.
The ever popular new-age database MongoDB topped the list again this year, with Redis, a tool for managing data, and ElasticSearch, which provides the foundations for building your own search engine, as runners up.
The lesson, both this year and last: if you want to gain traction in the database management system market, its a good idea to open source the thing, making the code freely available to the world at large.
The flagship product from software giant Oracle has dominated the database landscape for years, and it’s still sitting pretty at the top of the DB-Engines ranking, followed closely by MySQL, a database the company acquired as part of its purchase of Sun Microsystems in 2009. But the market has diversified in recent years, thanks to new data management techniques pioneered by the likes of Google and Amazon.
Drawing on these techniques, so-called “NoSQL” databases ditch the traditional relational database structure of storing data in neat rows and columns and enable developers to spread data across hundreds or thousands of servers—or develop much simpler applications that don’t as much structure. Oracle has tried to break into this field with its own Oracle NoSQL database and Oracle Big Data Appliance, but the real action remains in the world of open source, according to DB-Engines.
The DB-Engines ranking system was created by the Austrian technology consulting firm Solid IT to help its developers decide which new technologies were worth learning about and which were marginal. To rank databases, Solid IT analyzes data from many sources, such as Google Trends, LinkedIn, and various job listing sites, and assigns each system a score. To determine the fastest growing database systems, the company looked at which ones had the biggest change in their scores.
Closed source databases like Oracle, Microsoft SQL, and IBM DB2 are still enormously popular, according to DB-Engines, but the fastest growing databases are all open source. And according to a separate report, open source overall databases are gaining traction faster than closed source databases.
Last year’s biggest movers were MongoDB, the more traditional database PostgreSQL and Cassandra, a big data system open sourced by Facebook. This year, MongoDB is still growing faster than any other database system. It’s now the 5th most popular database overall.
Redis, which took second place, has long been the third most popular NoSQL database on DB-Engines, and its popularity is continuing to grow. Unlike systems like Cassandra and Hbase, which focus on juggling enormous amounts of data, Redis is designed to help developers handle small amounts of quickly changing data, such as whether a particular user is logged in or not, and what’s in their shopping cart.
But according to Solid IT co-founder Matthias Gelbmann, it’s starting to be used for more than just that. “We often see, that once Redis is installed for caching, and people see the speed and reliability of it, they start moving more and more functionality there,” he wrote in a blog post about the DB-Engines rankings.
Meanwhile, ElasticSearch provides another piece of the data puzzle: it makes it easier to actually find all the information that you collect. Not unlike closed source products from companies like Autonomy, ElasticSearch can be used to index internal documents and other files so that companies can create their own internal Google-style search engines.
A similar piece of software called Solr actually ranks higher on the DB-Engines list. But ElasticSearch, the newer of the two open source search engine platforms, had more growth in 2014. This suggests the two products might finally be on their way to displacing closed source alternatives.


Post a Comment