Commentary: It's nary longer breathtaking to speech astir the emergence of non-relational databases, but it's happening each the same.
It's not precisely wide wherever we are successful the Gartner Hype Cycle with respect to alleged "NoSQL" databases. We've decidedly been done the Trough of Disillusionment, but are we successful the Slope of Enlightenment oregon person graph, document, key-value and different specified databases deed the mainstream (Plateau of Productivity)? My consciousness is that we're astatine the beforehand extremity of mainstream adoption fixed however quiescent the media has gone connected databases similar Redis, Neo4j and MongoDB, contempt their rising popularity.
In different words, non-relational databases person ne'er been much popular, oregon much invisible successful that popularity. Until Neo4j's $325 cardinal money raise we'd each but forgotten databases astatine all, preferring to absorption connected shiny, caller options (Snowflake, anyone?). But it's precisely astatine this infinitesimal erstwhile databases person go uninteresting to speech astir that they're astir absorbing from an adoption perspective. Let's instrumentality a look.
SEE: Navigating information privateness (free PDF) (TechRepublic)
Tip of the information iceberg
Unfortunately, we don't person a batch of nationalist information to enactment with. Yes, we person MongoDB's gross information (up 39% year-over-year, with a rising percent shifting to cloud), but adjacent present we don't person the afloat story. We know, for example, that the institution present has implicit 25,000 customers, but this leaves retired the tens of thousands (or hundreds of thousands?) much that download the bundle and tally MongoDB successful their on-premises information halfway oregon self-host it with 1 of the unreality providers. MongoDB has reported that the database was downloaded implicit 70 cardinal times past year, which is adjacent much awesome erstwhile you see that it has been downloaded 175 cardinal times (cumulative) since 2007.
In different words, the popularity of MongoDB is accelerating. Fast.
Given the information successful DB-Engines' popularity rankings, MongoDB isn't unsocial successful this among the non-relational database crowd. We don't know, for example, what Neo4j's gross is, but we bash cognize its VCs saw capable committedness to money the biggest database task superior rise successful history. So let's look astatine the information that we bash have.
First, it's wide that relational databases proceed to dominate the information landscape, with Oracle, MySQL, SQL Server and PostgreSQL continuing to apical the charts. But this tells an incomplete story–it fundamentally tells america wherever the marketplace has been, not wherever it's going. If we instrumentality a antithetic presumption of the data, we spot non-relational databases spiking successful popularity, comparative to their relational database peers (Figure A).
This becomes adjacent much pronounced (and nuanced) if we look astatine conscionable the past 24 months (Figure B).
Keep successful caput that these charts measurement comparative popularity, not absolute. In presumption of implicit popularity, relational databases collectively garner 72.8% of the total. Graph databases? 1.7%. Time series, which shows specified a pronounced leap successful this past chart? Just 0.9%.
But papers databases, which are mostly represented by conscionable 1 database (MongoDB), travel successful astatine 10%. That's a important portion of the wide database pie, and increasing fast.
It's a spot similar cloud, successful a way. Cloud is increasing fast, but it inactive represents conscionable 5-6% of wide IT spending. Yet, arsenic IDC vice president Stephen Minton has noted, "Where determination is growth, astir of it is successful the cloud." The aforesaid is existent successful databases. Relational is wherever the marketplace is, but it's not wherever the maturation successful the marketplace is. That maturation is successful non-relational databases. Both are going to coexist for immoderate clip and astir apt forever. (After all, mainframes are ever with us. IT doesn't die–it conscionable stops increasing comparative to the remainder of the market.)
So, yes, the non-relational database marketplace is booming, and volition proceed to bash so. Minton, again: "The magnitude of information that companies indispensable store and negociate is not going anywhere. Increasingly, adjacent much of that information volition beryllium stored, managed, and progressively besides analyzed successful the cloud." The cloudier the data, the much it assumes the "three Vs" (volume, velocity, variety) of "big data," and overmuch of that keeps uncovering its mode into MongoDB, Prometheus, InfluxDB, Redis, Apache Cassandra and different non-relational databases. We don't needfully speech astir it, but it's happening, each the same.
Disclosure: I enactment for AWS, but the views expressed herein are mine, not those of my employer.
Data, Analytics and AI Newsletter
Learn the latest quality and champion practices astir information science, large information analytics, and artificial intelligence. Delivered MondaysSign up today
- Postgres is hot, but developers haven't mislaid their emotion for MySQL. Here's why (TechRepublic)
How to win successful bundle engineering management (TechRepublic)
How to go a information scientist: A cheat sheet (TechRepublic)
Big data's relation successful COVID-19 (free PDF) (TechRepublic download)
Power checklist: Local email server-to-cloud migration (TechRepublic Premium)
Big data: More must-read coverage (TechRepublic connected Flipboard)