Bob Muglia is something of a database bard, capable of unfolding vast stories about the evolution of technology.
This is what Muglia, former Microsoft executive and former CEO of Snowflake, did so on Wednesday morning during his keynote at the Knowledge Graph conference in New York.
His lecture topic, “From the Modern Data Stack to Knowledge Graphs,” brought together approximately fifty years of database technology in a new form.
The basic story is: five companies have created modern data analytics platforms, Snowflake, Amazon, Databricks, Google, and Azure, but these data analytics platforms cannot perform business analysis, including, most importantly, representing the rules that underpin compliance and governance.
“The industry knows this is a problem,” Muglia said. The five platforms, he said, representing “the modern data stack, have enabled a” new generation of these very, very important data applications. “However, “when we look at the modern data stack, and we look at what we can do effectively and what we can’t do effectively, I’d say the number one problem customers have with these five platforms is the governance.”
Muglia, who led the SQL Server business at Microsoft, among other feats during a 30-year career building databases, gave illustrations of business rules that data platforms cannot. model.
“So if you wanted to make a query to say, ‘Hey, tell me all the resources that Fred Jones has access to in this organization’ – that’s a tough query to write,” he said. “In fact, it’s a query that probably can’t run efficiently on a modern SQL database if the organization is very large and complex.”
The problem, Muglia said, was that algorithms based on structured query language, or SQL, can’t perform such complex “recursive” queries.
“There have been many generations of algorithms built that were all created around the idea of a binary,” Muglia said. “They have two tables with the key to join the two together, and 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 such as Fred Jones permissions, he said, “cannot be solved effectively with these algorithms, period.”
The right structure for business relationships, as opposed to data relationships, Muglia said, is a knowledge graph.
“What is a knowledge graph?” asked Muglia rhetorically. He offered his own definition of what can be a sometimes mysterious concept. “A knowledge graph is a database that models business concepts, the relationships between them, and the associated business rules and constraints.”
Muglia, now a member of the board of directors of a startup relational AItold the public that the future of business applications will be knowledge graphs built on top of data analysis, but with the particularity that they will use relational computing dating back to relational database pioneer EF Codd .
“Go back to the start,” Muglia urged. “What is the fundamental algorithmic capability of relational technology, what can be done with it?”
“If we look at what Codd was creating in the 1970s, the theorem said that relational algebra and relational queries are exactly equivalent in power of expression – so interesting! I always knew it was interesting, and now I know why.”
By using relational AI technology, in particular the technology developed by founder and CEO Molham Aref for query optimization, Muglia said, combined with knowledge graphs, it is possible to bring the same relational algebra to organizing business concepts.
“If we moved to relational knowledge graphs, we now move to the fundamentals of relational computation, and we issue relational computation instructions that are unordered and contain business rules and constraints.”
The work to “fundamentally redefine the way we can use relational algebra”, Muglia said, has been going on for about 20 years, but gained momentum in 2010 with the work of Aref, and within numerous universities and companies, with hundreds of articles published on the subject, he says.
“It has been an incredible cooperative effort between research communities around the world,” he said. “Materialize uses it fundamentally inside the materialization they do. LinkedIn has embraced some of that through a graph database they use called Liquid.”
The Relational Knowledge Graph introduces a new language, called Rel, although “SQL remains important,” Muglia said, “SQL isn’t going away,” as it serves as a sort of on-ramp to the new world of the Knowledge Graph. relational knowledge. .
“I can even say that SQL’s best days are ahead of it.”
Muglia foresees “incredible, incredible potential from a software development perspective, and all kinds of things flow from that,” he said of the relational knowledge graph. “As the technology matures – and I want to focus on the fact that it’s not mature yet – but as the technology matures, we’re going to see things that we can do with it that we don’t. couldn’t even dream.”
This includes a much greater use of machine learning in business analytics, as the name relational AI suggests. Also, the business model, Muglia said, will no longer be “just something we put on a whiteboard for an engineer to look at to write Java code.”
“The model,” Muglia said, “becomes the program, and so business analysts can get involved and make changes to data structures.”
“Think of the thousands of people involved who know the business – think about it!”
Right now, relational AI technology is “a pretty white glove,” he said. “We have a number of organizations that we’re doing limited testing with,” and the company hopes to “open it up for self-service” next year with “a broad developer beta so people can s ‘sign up and start using the system.” It’s the same trajectory, he said, as when he was running Snowflake in 2014 and releasing early versions of his code.
“We are at the start of a whole new era,” Muglia said. “It’s like the modern data stack in 2013, 2014 – that’s where we are in this lifecycle.
“And just as the modern data stack has revolutionized analytics, I have no doubt that knowledge graphs, and, in particular, relational knowledge graphs, will transform the way businesses operate.”
The Knowledge Graph conference is in its fourth year, having started life as a small affair in a ballroom at Columbia University in 2019. This year, after two years of virtual-only proceedings, the conference has moved on. transformed into a sprawling hybrid event, with dozens of panels as well as live sessions at the Cornell Tech campus on Roosevelt Island in New York. The program runs until May 6 .