How to build with IBM and MongoDB Enterprise Document Store

Manager, Db2 Tools Product Management, IBM

The IBM-MongoDB partnership continues to go like gangbusters. Announced in June 2018, our first version of MongoDB Enterprise Advanced is now giving way to our second version.

Why did we start on this path?

It all starts with our customers’ hybrid data management strategy. The need to embrace the proliferation of data that is creating new opportunities for businesses to better understand their customers, their industry and their own operations.

What do I mean by “proliferation?” Well, recent studies have suggested that customers have between 100 and 200 database platforms. Think about that. In order to understand the full business implications of data, many, most, or all of these sources may have to be queried. Examples of this data can include:

  • sales data
  • problem data
  • customer data
  • product data
  • purchasing data
  • financial data
  • contract data

I may want to answer the question: “What do I know about a specific customer?” This may lead to a query which executes against a specific customers sales, problems, products, contracts and revenue data stores as an example.  

Why open source matters

The Hybrid Data Management team within IBM Analytics has made a commitment to add value to our clients’ ability to manage and enable better AI, understanding and decision-making on large numbers of disparate data sources. That includes data based on open source platforms. Why do companies leverage open source? Because data environments are often initially selected by developers, and developers often go for the low- or no-cost option with prototypes that eventually find themselves in production.

You may have seen IBM embracing open source with BigSQL and Hortonworks. And we’re following along with our Db2 family federation solution set. We enable clients to leverage the Common SQL engine to execute singular queries against Db2 and many disparate data sources.

And since June, we’ve added MongoDB into our platform offerings.

What’s new in V2

The IBM Data Management Platform for MongoDB V2 announcement delivers key Mongo Enterprise 4 capabilities into the market. These include:

  • Multi-document transactions against replica sets. With these transactions, no write operations are visible outside of the transaction until the transaction commits
  • Aggregation extensions adds a host of new aggregation operators for type conversion. This includes support for Boolean, Date, string, and numeric conversions              
  • Tooling extensions helping manage and grow public and private cloud deployments
  • Continuing security enhancements
  • And more              

To net it out: MongoDB continues to grow into the document store addressing operational requirements with JSON data. Read the announcement to learn more.

One of the most exciting capabilities added to this new release now includes the ability for customers to use the Mongo Documents Store Database with applications needing ACID level database assurance. (ACID stands for atomicity, consistency, isolation, durability). It’s essentially a set of definitions that define support for transactionality in database processing. With this support, our clinets can leverage MongoDB as a serious contender when modernizing existing applications as well.

Companies leverage transactional support when their applications need confirmation that multiple query updates happened and are confirmed. They then have a view for example of the order data being in synch with the product shipment data and the customers payment data as well. This level of assurance is required by many mission-critical application models.  

In addition, we see organizations continue to standardize on MongoDB as part of initiatives that include addressing operational requirements, IOT, and a myriad of other application models. And our Db2 family, governance family, and business analytics family of offerings, along with IBM Cloud Private integrate with MongoDB as well.

Customer requirements for private clouds

Many organizations have data environments that cannot leave the premises or a specific geographical location. As such, while these customers want to provide managed cloud solutions to their internal customers, they need supporting infrastructure to deliver that. Technologies like Kubernetes and Open Stack play a big part in easily spinning up environments, as well as managing and securing them. Once extended from the environment, they need processes to manage the influx of corporate data. Catalogs, governance, glossaries and secure access needs to be considered. And in parallel to management of the data, they need tools that help them deeply understand new data models, including visualizations and scoring.

Moving forward, IBM’s MongoDB Enterprise will fit into IBM Cloud Private and Cloud Private for Data, improving access to – and understanding of – the Json document data MongoDB Enterprise manages. Stay tuned for more on this topic later this month. In the meantime, read the IBM Data Management Platform for MongoDB datasheet.