Graph databases and analytics systems have been quiet, "under the covers" technologies for some time. Based on...
mathematical notions of structure, they did a pretty good job of modeling relationships between data elements in special settings, but their use has been narrow.
In terms of visibility, they certainly lag the relational database, which, by some measures, is actually younger than the graph. Now, graph technology is coming into greater prominence, as a driver of many new application architectures. It was at the crux of new products described last month at Informatica World 2016 in San Francisco.
The graph that's most familiar is the ubiquitous Facebook friends' network. Similar technology is behind Google initiatives, such as Knowledge Graph. A graph database underpins the Panama Papers project -- the International Consortium of Investigative Journalists' effort that made headlines when it disclosed the sundry offshore businesses connections of some world leaders.
There is more. Graph technology buttresses a semantic data system for delivering better patient diagnoses at the Montefiore Health System, an academic medical center and University Hospital based in Bronx, N.Y.
The general notion of graphs of information even showed up as a factor in Microsoft's $26.2 billion purchase of LinkedIn. Microsoft CEO Satya Nadella cited LinkedIn's professional network graph as one of the jewels of the acquisition, one he hopes to closely connect to the Microsoft Graph of customers' calendars, project artifacts, documents and so on.
Where relationships need careful tracking, homebrewed graph databases are increasingly common. Master data management (MDM) has been one such use. MDM software vendors use graph technology in various forms, as well, with the objective to improve their MDM and data governance offerings. That's seen in recent releases of Cambridge Semantics' Anzo graph software for data management, as well as Reltio's Cloud and TopQuadrant's TopBraid platform.
Graph technology takes center stage
The drum roll of this type of activity is not lost on data integration vendor Informatica. If not exactly front and center, graph technology is the beating heart within several new Informatica products, such as Live Data Map.
In effect, Live Data Map acts as a knowledge graph and metadata repository, and can help automation of data discovery and preparation tasks. The software is still new, having reached version 1.0 in December.
At the conference, it was discussed as an underlying technology behind Informatica's upcoming Enterprise Information Catalog. It has been used to support the company's Big Data Management framework, as well as its Secure@Source, Intelligent Data Lake and PowerCenter product lines.
"Live Data Map is baked into everything we are doing. But it is not just about connecting Informatica products; it's about connecting data throughout the whole enterprise," Amit Walia, executive vice president and chief product officer at Informatica, based in Redwood City, Calif., said in an interview. "This is the central nervous system, or the brains of a company. It becomes the Google of the enterprise," he continued.
MDM is a possible use of Live Data Map, according to David Gleason, a head of data strategy and governance at New York-based JPMorgan Chase, who took part in a conference panel on data management and big data analytics.
"We are looking at the Live Data Map tool as a way to [view] and make inferences about data. For me, right now, it is an experiment or a work in progress, but it definitely represents the direction we are going in," he said.
Focus on innovation
Informatica company leaders pointed to Live Data Map as one among several signs of the company’s commitment to innovation. They indicated the company has worked to improve performance of the open source Titan graph database on which Live Data Map, originally discussed at last year's edition of the conference, was built. Better performance is often at issue with such systems; it has, in fact, been limited the growth of the graph of data in the past.
Live Data Map will help Informatica implement its larger innovation roadmap, according to Stewart Bond, who is director of data integration software at IDC.
"I am excited to see what Live Data Map has come to be. It's about the unification of metadata," he said.
Better metadata repositories will gain added importance as unstructured data engulfs organizations, according to Bond, who said he anticipates Live Data Map's capabilities to help in this regard. This and other efforts are closely watched, given changes in Informatica's overall corporate structure.
This year's Informatica World was the first since the company went private. In 2015, it executed a $5.3 billion leveraged buyout backed by venture capitalists, pension funds and no less important software industry players than Microsoft and Salesforce. Buyout moves like this are often scrutinized for signs that new product development efforts are getting short-changed.
That doesn't seem to be the case here, IDC's Bond said. He suggested Live Data Map and other technology activity at Informatica World 2016 show a company still very much in the game.
"They didn't 'go quiet,' they didn't get broken up and, at the same time, we have seen the company bring forward new innovations," he said.
Learn the ins and outs of a GPU database for analytics
Find out how graph databases fit with other NoSQL types
Discover graph data's role in an accommodation platform
AWS enters an area of graph database technology