Imply, the startup founded to productize the Apache Druid database in pursuit of analytics in motion, this week announced that the completion of a $70-million Series C round of funding.
Imply was founded in 2015 by some of the creators of Apache Druid, a column-oriented in-memory data store that was originally developed at Metamarkets, an ad tech analytics firm. The goal with Druid, which was created in 2011, was to serve OLAP-style analytic queries on high-dimensional data at low latencies, something that the Hadoop-based distributed processing projects (Hive, Impala, and later Presto) of the day struggled to do.
In this way, Druid succeeded in marrying the analytical capabilities of traditional column stores, such as Vertica, Netezza, or Greenplum, with the speed of in-memory processing. After the core engine was developed, streaming data capabilities were added to the mix, resulting in the “shapeshifting” (or Druidic) capability to change from processing queries rapidly from streams of historical data arriving from batch sources (HDFS and S3 data lakes) as well as the freshest fast-moving data from modern message busses (like Apache Kafka and now Amazon Kinesis).
Like other commercial open source vendors, Imply is building a cloud platform that simplifies management of the distributed analytic product, thereby enabling customers to focus on using Druid to crunch the fast-flowing data to benefit their businesses. According to Imply CEO and co-founder Fangjin Yang, Imply will use the infusion of venture funding to bring this business model to fruition.
Our vision at Imply has always been to create a new category for data analytics, analytics-in-motion, and enable organizations to unlock workflows they’ve never been able to do before, Yang wrote in a blog post yesterday.
We’ll be soon rolling out Imply SaaS, which will include significantly expanded and simplified data modeling, data ingestion, and query optimization capabilities.
When the folks at Imply started developing Druid, the plan was to enable analytics
at the speed of thought, Yang said. The core features of Druid enables features like observability analytics, network performance analytics, fraud operations, digital marketing analytics, clickstream analytics, and supply chain analytics, he wrote.
We still learn about new use cases every day, but what ties all the applications of Druid together is that every use case utilizes analytics-in-motion, he wrote.
End users aren’t just looking at a static dashboard, they are actively engaging with data by asking questions, obtaining responses, and using those responses to ask further questions.
The Imply team has delivered a number of capabilities over the past few years, according to Yang, including Kubernetes support, exactly-once streaming ingestion, development of a new vectorized query engine, dependency-free batch ingestion, SQL support, table-level security, row/column-level security, and numeric column support. It created a data visualization product called Pivot, a monitoring tool called Clarity, and a deployment and security monitor, Yang said.
I couldn’t be more proud of the Imply team and what we’ve accomplished to date, Yang wrote.
The Series C is a major milestone in our journey, but we’re just getting started. We have ambitious plans of where we want to take the product and the road ahead is sure to be filled with fun and difficult challenges.