When I first met Gary Bloom more than 15 years ago he was second to Larry Ellison at Oracle, and when he jumped to run Veritas, I was one of the first reporters to interview him. Then came a short stint at Symantec, eMeter, then consulting with private equity firm TPG. With over 30 years in key positions he's become one of Silicon Valley's top business leaders.
Since 2012, Bloom has been CEO of MarkLogic, a fast-growing Silicon Valley database company riding the hot wave of interest in NoSQL databases such as MongoDB, as large organizations attempt to mine their diverse stores of big data for profitable business insights.
Relational databases still run at the core of every large business but they are seen as too limited and too structured for the digital business world where there is a need for highly scalable, highly adaptable database architectures that can easily add unstructured data sources, run on clusters, and support complex web sites -- without first mapping all the data into the correct rows and columns of a relational database.
A NoSQL database doesn't do away with SQL -- the standard query language used by relational databases, it simply means "not just SQL" -- users can query the data in a variety of ways. Structured Query Language (SQL) is hard to use and requires detailed knowledge about the data, requiring highly trained specialists to run reports. MarkLogic offers an alternative to SQL, a semantic technology that Bloom says can understand queries in natural language sentences. Anyone can interrogate vast stores of data without needing data scientists and SQL specialists.
It might seem strange to group MarkLogic, founded in 2001, with the same hip, young crowd of NoSQL databases such as MongoDB, Cassandra, Couchbase, Hadoop, MapReduce, HBase, Hypertable and many more. But MarkLogic fulfills the definition of a NoSQL database and it also has something the others don't: enterprise-ready features.
And Bloom's long experience in enterprise IT means that he knows how to speak to large corporations, he knows what they need, what scares them, and he knows what isn't working.
"What isn't working is the data integration projects that large companies have tried to do with relational databases. Many of those projects have become stalled because of the limitations of rows and columns. We can handle both unstructured and structured data and that means integration is faster and far less costly."
Searching for data...
MarkLogic was founded by Christopher Lindblad, a former architect at InfoSeek -- one of the first search engines. His goal was to apply search-based approaches to indexing data and searching metadata. MarkLogic is document-centric, which is why media companies were early adopters because they could use it as a very sophisticated content management system.
Media companies are still buying MarkLogic, customers such as NBC, which used it to create a highly popular app for streaming Saturday Night Live comedy video clips. Two sectors: healthcare and financial services are the most important sectors today.
Business has been very good Bloom says, with annual sales topping $100 million. The driver is Big Data: large corporations are convinced there is an El Dorado of untapped commercial opportunities -- if only they can run their reports across all their data sources. But integrating all that data is too costly, and takes too long with relational databases. The future will be full of data in many forms, formats, and sources and how that data is used will be the differentiator in many competitive battles. If that data can't be searched it can't be used.
Companies have been able to easily test NoSQL databases for data integration projects because they are open source software or there are free developer versions. The results have mostly been good but moving to the next stage of deployment is not so easy because of the lack of enterprise-ready features. This is the MarkLogic differentiator over its younger competitors.
"When you use Oracle, for example, you don't have to worry about security because it is built-in.You shouldn't have to worry about enterprise issues such as implementing policies," says Bloom. "MarkLogic is the only NonSQL database that is enterprise ready. And we are also the only transactional NoSQL database."
Bloom knows how to speak to risk-averse IT organizations, and he knows what keeps CIOs and CEOs awake at night. One of their biggest challenges he says, is all the data spread around the organization and held in silos. This obstructs companies from generating a single view of their business operations and discovering valuable insights that can be turned into new revenues.
MarkLogic v Oracle...
The timing looks perfect for MarkLogic, it has the right leadership and its technology is in the right place to capitalize on the pursuit of Big Data. Its revenues have grown tremendously as some of the world's largest companies seek not only Big Data insights but also some relief from the international complexity of regulatory compliance across all their business groups. Relational databases cannot provide the data agility that they need to do what they want.
Although MarkLogic is tiny compared to Oracle there are some interesting parallels. "MarkLogic is at about the same size as Oracle was when I began working there. It took a long time for Oracle to get security and other enterprise features right, but when it did, that was when company really took off."
Oracle rode the industry-wide move away from mainframes in the 1990s, to become a giant in the IT sector. These days it is losing deals to MarkLogic Bloom claims. The government's botched Healthcare.gov launch was corrected by moving from Oracle to MarkLogic, and Bloom says many other customers have had to pull the plug on Oracle based projects.
Another similarity with the early Oracle is MarkLogic's work for government spy agencies, a booming business for Silicon Valley's Palantir Technologies -- a venture-backed startup valued at about $20billion.
"Unlike Palantir, which brings in data scientists for a single project, we sell systems that enable agencies to analyze all their data sources in real-time, all the time, with real-time notifications -- not just one-time projects," says Bloom.
Is it still possible to build a large tech company such as the next Oracle in Silicon Valley, when the IPO market continues to be weak, and the huge incumbents have global platforms that will take years for challengers to match? We will have to wait and see.
MarkLogic is one of the few Silicon Valley companies that has the technologies, the positioning, and the revenues to have a successful IPO once the markets return. The IT department is changing rapidly, it is being re-organized, re-tooled and re-imagined so that it can support a fast moving digital business.
MarkLogic is betting on IT becoming strategic again, that IT matters once more. It's a very good bet.