Your Resource for All Things Apps, Ops, and Infrastructure


At the beginning of October, Cloudera and Hortonworks announced that they would be merging in a deal worth $5.2 billion.

At the time, Cloudera CEO Tom Reilly, stated, “Our businesses are highly complementary and strategic. By bringing together Hortonworks’ investments in end-t-end data management with Cloudera’s investments in data warehousing and machine learning, we will deliver the industry’s first enterprise data cloud from the Edge to AI. This vision will enable our companies to advance our shared commitment to customer success in their pursuit of digital transformation.”

Cloudera’s shareholders approved the Hortonworks merger in a vote that was held on December 28th, where 112.8 million shareholders voted in favor with 134,000 votes against, and 396,000 abstaining. Shortly thereafter, Hortonwork’s shareholders approved the merger, although by a smaller margin, with 60 million voting in favor and one million against. As a result, the all-stock merger officially closed on Thursday last week.

New Cloudera

The newly merged entity is called Cloudera and will be led by Cloudera’s former CEO, Tom Reilly. As part of the combination, many executives have been reshuffled and eleven integration groups were formed to smooth the transition.

Arun Murthy, co-founder and Chief Product Officer of Hortonworks and now Chief Product Officer of the new Cloudera, says, “Our primary initiative out of the gate is to deliver a 100-percent open-source unified platform, which leverages the best features of Hortonworks Data Platform (HDP) 3.0 and Cloudera’s CDH 6.0. Cloud-native and built for any cloud – with a public cloud experience across all clouds – the unified platform embodies our shared “cloud everywhere” vision.”

What Led to the Merger

Hadoop has changed significantly since the days when Cloudera was founded as the first Hadoop service provider back in 2008. Cloudera was a small ecosystem, and Hadoop was comprised of HDFS and MapReduce. At that time, Hadoop was founded on the premise of processing the data where it already sits on-disk because it was faster to read from a disk on a local commodity system than to read from a SAN and transmit over a network.

With cloud computing, we have broken that paradigm. Due to the benefits of cloud, we now opt to store the data in a centralized location (e.g., S3 and Cloud Storage) and transmit over a high bandwidth connection to process the data elsewhere. By using solutions like S3 or Cloud Storage, customers have decoupled the very premise on which Hadoop was founded. Now, customers are running separate frameworks, such as Presto or Impala, that can be built without MapReduce.

The merger of Cloudera and Hortonworks is an acknowledgment to the commoditization of Hadoop as a whole – with each entity selling support and add-on services to the ecosystem.

The cloud was pivotal in the merger. There’s no question that the growing adoption of AWS, Azure, and Google Cloud services comes at the expense of on-prem infrastructure and software. These major cloud providers all have competitive Hadoop offerings, which include support offered as part of the standard contract. The cloud providers’ Hadoop offerings are stripped down compared to the established Hadoop vendors Cloudera and Hortonworks. However, now with plenty of quick-start guides for running your first word-count job in under an hour, there is a very low barrier to entry.

Unified Future

Public cloud services are predicted to continually increase over the coming years. For reference, Gartner believes they will grow 18%, to become a $247 billion business worldwide – with the cloud accounting for the majority of analytics purchases by 2020. It appears Cloudera and Hortonworks recognized that together their offerings could work both in the cloud, across all major public cloud infrastructures, and on-prem. As part of the merger, the new Cloudera plans to provide an end-to-end hybrid data platform that spans customers’ data centers, public cloud environments, and Edge.

Subscribe to the AHEAD i/o Newsletter