Forbes: How Graphs & NoSQL Power Advanced Data-driven Apps
In the modern world of data-driven applications we are at a fascinating point at which both fully formed products and powerful components are being offered to us at a breathtaking pace.
The question that companies of every size must answer is: How can I make use of these products and components to build apps to improve my business?
One way is to buy fully formed products and configure them to meet your needs. This is what most companies do outside of Silicon Valley and the pockets of industries where rapid and continuous innovation is the key to success. Another way is to use components to create custom solutions. This is the approach that is taken by the industry leaders who seek to disrupt. You can’t just buy that disruptive app off the shelf, now, can you?
But there is a third way, and that is to be inspired by products but assemble a solution that imitates a pattern shown by a product. This approach has significant promise and I would like to explore it using Reltio as an example. Reltio is a product that combines Master Data Management (MDM) with the power of the Cassandra NoSQL and graph databases to create a new kind of application development platform for data-heavy use cases. (Reltio refers to itself as a “platform-as-a-service” vendor, but the point of PaaS is to enable the creation of applications, so I prefer the term application development platform.)
The Anatomy of Reltio
Reltio does something that few companies have done as an enterprise software product: create an application development platform that exploits the power of graph technology. Neo Technology‘s Neo4J graph database has shown the power of graphs as a component for powerful applications. Teradata’s Connection Analytics platform is able to rapidly assemble graphs that can then be subject to powerful analytics.
Reltio’s first insight is that graphs are most powerful when they reveal new relationships that spring from a combined, curated set of data. As it happens, a body of practices for combining and curating data has developed over the last quarter century as part of the implementation of data warehouses. That body of practices is called Master Data Management (MDM). Reltio realized that to make powerful graph applications, you always start with MDM, and so the foundation of Reltio is an MDM engine.
But Reltio realized that its MDM engine needed to be implemented in a new way to meet the needs of both a modern software product and the arrival of big data.
In the big data era we’re now in, companies have to be able to synthesize data from an immense amount of sources, both internal and external. In today’s data landscape, for instance, customer information can arrive in a mind-boggling number of forms and from an equally large variety of sources — loyalty service, transaction, shipping, social media, and buying history can all be located in different places. To fully understand a customer, you need to create a record that combines all of that information. And this is as true for products as it is for people. All your data needs to be unified.
An effective MDM platform is thus essential for getting the full narrative of your data, as it allows this type of cohesion to take place. Many MDM systems have struggled with how to do this gathering and get the data into a form that’s actionable.
Instead of basing its MDM and other data management capabilities on relational databases, which is the most common choice, Reltio loads the combined and curated data into Cassandra, the NoSQL database developed by Facebook. This allows two things. First, it provides a way that Reltio can provide a scalable multi-tenant platform-as-a-service to implement a modern data management platform. Second, it allows clients to easily support huge datasets. In addition, Cassandra easily accommodates a rapidly evolving schema, something most relational databases find difficult.
But the key is that MDM is just the precursor for what matters most with Reltio – graph analysis. MDM unleashes the strength of data-driven applications because it provides a true 360-degree view. This potent MDM engine helps to create the foundational data objects for analytics to take place. Graphs are the best way to see relationships between people, data, and things. And this is where Reltio’s true value emerges – the graph semantics and analytics that can be applied on top of the foundational data engineering.
The power of graphs cannot be overstated, and they grow more formidable as companies incorporate various types of semantics to the graphs. For example, for a pharmaceutical company, public committees often approve new drugs. Reltio’s graphs can be used to alert a sales team when a new committee member is appointed and then show all the personal connections that the business has to that individual to initiate a reach-out. Or in the case of manufacturing, component inter-dependencies can be depicted through a graph so that a company can be warned if a particular item fails and see how it will affect the entire pipeline of work.
Another important aspect is that data-driven applications from Reltio and their partners are both analytical and operational. They provide not just insights and recommended actions, but allow business users to take actions through social data accessible across an enterprise.
Productization of Data-driven Graph Apps
Reltio is methodically discovering semantic patterns that provide ways to recognize key events and to suggest or automate intelligent actions and responses in a variety of domains. In this way, Reltio is productizing the power of graphs to build data-driven applications. Reltio allows apps to be created through a process of configuring MDM, the graph, and the infrastructure for event recognition and response. As more data is added, more events can be recognized and new responses can be added through a combination of configuration and custom coding if needed.
Companies could get this type of data and analytic insight on their own, using MDM and building a graph database on top of it. And for some companies, that may make sense. But for companies that don’t have a fleet of MIT-trained engineers on staff, Reltio offers a full-spectrum data management platform-as-a-service incorporating a hybrid columnar graph database and analytics in a variety of domains at an affordable price. With Reltio, companies avoid accruing a large technical debt to get a great result. Reltio will be a way to create applications that deliver significant value and also create a formidable institutional memory. It’s an interesting example of where data technology is going in the future.
Follow Dan Woods on Twitter
Dan Woods is on a mission to help people find the technology they need to succeed. Users of technology should visit CITO Research, a publication where early adopters find technology that matters. Vendors should visit Evolved Media for advice about how to find the right buyers. See list of Dan’s clients on this page.