Microsoft veteran Bob Muglia: The Relational Knowledge Graph will transform business

Bob Muglia is a database geek, capable of uncovering insightful stories in the evolution of technology.

That’s what Muglia, former Microsoft executive and former Snowflake CEO, did Wednesday morning during his keynote at the Knowledge Graph Conference in New York.

The subject of his talk, “From the Modern Data Stack to the Knowledge Graph,” unifies nearly 50 years of database technology in a new form.

The basic story is this: Five companies have created modern data analytics platforms, Snowflake, Amazon, Databricks, Google, and Azure, but those data analytics platforms can’t do business analytics business, including, most importantly, representing strict compliance and governance rules.

“The industry knows this is a problem,” says Muglia. The five platforms, he said, represent “the modern data stack, which has enabled a “new generation of these very, very important data applications” to be built. modern data stack, and we look at what we can effectively do and what we can’t do, I’d say the number one problem that customers are having with both These five foundations are governance. ”

Also: Ethernet creator Metcalfe: Web3 will have all kinds of ‘network effects’

Muglia, who runs the SQL Server business at Microsoft, among other feats in his 30 years of database building career, provides illustrations of business rules that data platforms don’t. modelable.

“So if you want to do a query to say, ‘Hey, tell me all the resources that Fred Jones has access to in this organization’ – that’s a difficult query to write,” he said. speak. “In fact, it’s a query that may not be able to execute efficiently on any modern SQL database if the organization is very large and complex.”

The problem, says Muglia, is that algorithms based on the structured query language, or SQL, cannot perform such complex “recursive” queries.

“There have been generations of algorithms built around the idea of ​​a binary system,” says Muglia. “They have two tables with a key to join the two tables together, then you get a result set and the query optimizer takes and optimizes the order of those joins – binary join , binary join, binary join!”

Recursive problems like Fred Jones’ rights, he said, “can’t be solved efficiently with those algorithms.”

The right structure for business relationships, as distinct from data relationships, is a knowledge graph, says Muglia.

“What is a Knowledge Graph?” Muglia asked rhetorically. He gives his own definition for what can be a sometimes mysterious concept. “A Knowledge Graph is a database that models business concepts, the relationships between them, and the related business rules and constraints.”

Muglia, currently a board member for the startup AI relationtold the audience that the future of business applications would be knowledge graphs built on top of data analytics, but with a twist that they would use relational calculus going back to the first company. EF Codd relational database style.

“Back to the beginning,” Muglia urged. “What is the fundamental algorithmic capability of relational technology, what can be done with it?”

“If we look at what Codd created in the 1970s, the theorem says that relational algebra and relational queries are exactly equivalent in terms of expression power – interesting! I’ve always known this! very interesting, and now I know why.”

Using Relational AI’s technology, especially that developed by founder and CEO Molham Aref for query optimization, combined with knowledge graphs, can bring relational algebra, says Muglia similar to the organization of business concepts.

“If we switch to the relational knowledge graph, we now come to the fundamentals of relational calculus and we give unordered relational analytic statements and they contain professional rules and constraints.”

Work on “fundamentally redefining how we can use relational algebra,” says Muglia, has been going on for about 20 years, but began successfully in 2010 with Aref’s work, and inside many universities and companies, with hundreds of articles published on the subject, he said.

“It is an incredible collaborative effort among research communities around the world,” he said. “Materialize uses it in a fundamental way inside the Materialization they do. LinkedIn has applied some of this through a graph database they use called Liquid.”

The Relational Knowledge Graph introduces a new language, called Rel, although, “SQL is still important,” says Muglia, “SQL is not going away,” as it serves as a kind of conduit to New world of relational knowledge graph.

“I could even argue that SQL’s best days are ahead”.

Muglia foresees “awesome, amazing potential from a software development perspective and all that comes out of this,” he said of the relational knowledge graph. “As technology matures – and I want to focus on the fact that it is still immature – but as technology matures, we will see things we can do with it that we couldn’t even dream of. next.”

That includes a greater use of machine learning in business analytics, as relational AI is called. In addition, the business model, Muglia said, will no longer be “something we put on the whiteboard that an engineer needs to look at to write some Java code.”

“The model,” says Muglia, “becomes the program and thus the business analysts can join and make changes to the data structure.”

“Think of the thousands of participants who know about the business – think about it!”

For now, Relational AI’s technology is “a nice white glove,” he said. “We have a handful of organizations that we’re doing limited testing,” and the company hopes to “open to self-service” next year with a “wide developer beta so anyone can post.” sign up and start using the system.” It was the same trajectory, he said, as when he ran Snowflake in 2014 and it released the first versions of its code.

“We are starting a whole new era,” says Muglia. “It’s like the modern data stack in 2013, 2014 – that’s where we’re at in that lifecycle.

“And just as the modern data stack has revolutionized analytics, I’m absolutely convinced that knowledge graphs, and specifically relational knowledge graphs, will change the way businesses work. .”

The Knowledge Graph conference is in its fourth year, starting life as a small affair in a ballroom at Columbia University in 2019. This year, after two virtual sessions, the conference has become a reality. Brilliant mixed event, with dozens of panels as well as live sessions at the Cornell Tech campus on Roosevelt Island in New York City. The program lasts until the end of May 6 .