What Are NoSQL Databases?

NoSQL open source Why the Modern Enterprise Needs NoSQL

Why the Modern Enterprise Needs NoSQL

Traditionally, enterprises would store data in SQL-based relational database management systems (RDBMS). All of that data was structured data—like financial statements, personnel files, product information and company records.

While SQL-based RDBMS are very capable pieces of technology in many ways, the proliferation of the internet and mobile devices has led to a surge in unstructured or unorganized data which RDBMSs can’t accommodate easily.

To meet the needs of the modern enterprise that generates enormous amounts of unstructured data in the age of big data, NoSQL databases have emerged.

Despite how the name might sound, these databases weren’t designed to displace SQL databases. In fact, some NoSQL databases—like OrientDB—actually support SQL language. A better way of thinking about NoSQL is that it’s the database designed to get developers and business professionals to “open their minds and consider new possibilities beyond the classic relational approach to data persistence.”

Whereas traditional RDBMSs are schema-oriented—which means that data stored in them needs to be structured or classified in way that adheres to the database’s “rules”—NoSQL databases can process unstructured data (e.g., blogs, videos, pictures, audio files and texts). In the age of big data, when organizations are collecting more and more data than ever before, this is no small feat.

Quite simply, NoSQL databases enable organizations to unlock the full potential of their unstructured data. It’s the reason why today’s leading enterprises are increasingly relying on them. In fact, a recent study predicts the NoSQL market will grow to $4.2 billion by 2020.

The Benefits of NoSQL

So what, specifically, sets NoSQL apart from its RDBMS counterparts?

  • Scalability. NoSQL databases are designed to process big data. To do this, they need to be able to scale out efficiently and significantly.
  • Performance. Unlocking the full potential of your data requires speed, which NoSQL databases are known for. Many of these databases can also process data in real time—which SQL databases simply cannot.
  • Flexibility. In SQL databases, data is typically stored in table format. In NoSQL databases, data can be stored in a number of different ways, including in graphs, documents and key-value, geospatial and reactive models.
  • Simplicity. NoSQL databases easily scale out horizontally. Scaling SQL databases—while not impossible—is a much more difficult endeavor.

The Challenges of NoSQL

While NoSQL boasts a number of benefits, it is not without its difficulties. Still, partnering with the right NoSQL vendor can help you overcome these challenges with ease.

  • Support. NoSQL databases tend to be open source projects. When something goes wrong, you may have luck reaching out to the community of developers and finding your answer. Unfortunately, today’s strongest enterprises can’t afford to roll the dice on products that are less-established and lack comprehensive support. Good news: Today’s leading NoSQL companies tend to offer free commercial products as well as premium products designed for organizations which tend to include enterprise-level support.
  • Maturity. While NoSQL has been around for quite some time, it wasn’t until recently that NoSQL databases started making serious inroads into the modern enterprise. As such, many of the NoSQL solutions that are out there aren’t as mature as many organizations would like—particularly when compared to their RDBMS counterparts. If you’re looking for a mature NoSQL database, look no further than OrientDB, which was the first multi-model database released—way back in 2011.
  • ACID compliance. Companies, and e-commerce organizations in particular, care a great deal about what’s called ACID compliance. Databases are said to be ACID-compliant when they can ensure atomicity, consistency, isolation and durability in each transaction. Essentially, ACID compliance ensures that databases process transactions as expected. While a number of NoSQL databases aren’t ACID-compliant, several—including OrientDB—are.

Customers who use OrientDB Enterprise Edition get access to enterprise-grade support while using the world’s most established multi-model database that is ACID-compliant.

NoSQL Models

Whereas SQL databases usually store data in table format, NoSQL databases—which are much more versatile by design—open themselves up to multiple formats. Some of the more common data models found in NoSQL databases include:

  • Key/value databases. Data models are reduced to tables consisting of key/value pairs that are often distributed across servers. 
  • Column-oriented databases. In these databases, data is stored in columns to allow for flexibility and aggregation.
  • Document databases. Here, the data model consists of document collections. Each document can have multiple fields without needing to have a defined schema.
  • Graph databases. Like the name suggests, graph databases store data in a model that consists of vertices interconnected by edges.