Hortonworks—the big-data company spun off from Yahoo—has confirmed that it will make its initial public offering on Friday. The company hopes to raise $100 million from 6,250,000 shares offered at $16 per share under the NASDAQ symbol HDP. Underwriters will have the option of purchasing an additional 937,500 shares.
Hortonworks sells support and services for Hadoop, an open source number crunching platform inspired by work first published by Google in the early 2000s. The software—whose development is overseen by the non-profit Apache Software Foundation—has spawned several different companies — including Cloudera and MapR — and Hortonworks is the first of these to go public.
The Hortonworks IPO is good news for all of these startups, says former Yahoo CTO Raymie Stata, who founded a startup dedicated to offering Hadoop in the cloud called Altiscale. “It’s a confirmation of what most people already think about Hadoop: that it’s becoming a must-have piece of infrastructure,” he says. “And that open source is becoming the preferred form for IT buyers.”
It should be said that Hortonworks stomached a $86.7 million loss on $33.3 million in revenue so far this year, according to an Security Exchange Commission filing made public last month. But the technology it offers is proven. Hadoop has gone from being an obscure tool used by web companies like Yahoo, eBay, and Facebook to powering large scale data analysis in even stodgy old industries like financial services.
Hortonworks co-founder and architect Arun C. Murthy says that Hadoop’s success was about much more than Yahoo or Hortonworks. It was about the whole community of contributors, both inside and outside Yahoo. “You had Facebook come in, LinkedIn wanted to do social graph analysis,” he says. “Hadoop became a great amalgamation of all these use cases. It was improved in all these directions simultaneously by all these different contributors.”
And these collaborations spawned much more than just Hadoop. LinkedIn spun-out Confluent to commercialize its open source tool Kafka. Former Facebook engineer Jonathan Gray started Continuuity to offer a cloud service for storing and processing data inspired by his former employer’s solution. University of California computer scientists Matei Zaharia and Ion Stoica started Databricks, a company based on their next-generation data crunching platform Spark.
“I hope we look back in five years and say this IPO was just the beginning,” Stata says.