News

MapR Blasts Open Data Platform

• By David Ramel
• April 23, 2015

MapR Technologies Inc. declined to join the young Open Data Platform (ODP) and today penned a blog post explaining why.

"MapR was invited to participate in the ODP initiative and declined after carefully considering the value to the market place," exec John Schroeder said today.

The ODP was formed in February by 15 industry organizations "to promote open source-based Big Data technologies and standards for enterprises building data-driven applications." Along with MapR, Cloudera Inc. also chose to not participate in the initiative.

Just last week, the consortium announced it had achieved one of its stated goals: to develop a standard "ODP core" that organizations could use to base projects on.

ODP members Hortonworks Inc., IBM and Pivotal Software Inc. said their respective Big Data products are now aligned on this core, based on version 2.6 of the open source Apache Hadoop software. Hortonworks, along with Cloudera and MapR, is generally recognized as one of the top three Hadoop distribution vendors, who have had a somewhat contentious relationship.

That relationship continued with today's MapR blog post.

"The announced ODP benefits Hortonworks marketing and provides a graceful market exit for Greenplum Pivotal," Schroeder said. "Should the ODP charter change in the future, we might consider participation."

Pivotal helped form the ODP while announcing it was open sourcing its Big Data platform, including the Pivotal Greenplum Database.

MapR provided many reasons for its refusal to participate in the ODP, saying it was redundant with Apache Software Foundation (ASF) governance, which is based on a "meritocracy."

MapR further stated the ODP core is solving problems that don't really exist, such as vendor lock-in and interoperability.

The company also had issues with the ODP core, which it characterized as "misdefined" and "vendor-biased." Schroeder said that while some components of the MapReduce/YARN/Ambari/Hadoop Distributed File System (HDFS) core are in wide use, other notable projects are gaining market share, such as Spark and Mesos.

The Apache Ambari project for Hadoop management is championed by Hortonworks, its most active developer.

"Ambari is used by less than 25 percent of the market," Schroeder said. "Hadoop was architected to support plug-and-play alternative technologies to HDFS. HDFS was built to serve as secondary storage for batch Hadoop processing. Many production use cases requiring POSIX-compliant storage replace HDFS with MapR, IBM GPFS, EMC Isilon or NetApp."

Also, Schroeder claimed that Hadoop leaders aren't participating in the ODP.

"[About] 75 percent of Hadoop implementations run on MapR and Cloudera," Schroeder said. "MapR and Cloudera have both chosen not to participate. The ODP without MapR and Cloudera is a bit like one of the Big Three automakers pushing for a standards initiative without the involvement of the other two.

"The ODP is not open unless equal voting rights are provided to the leading Hadoop distributions," Schroeder continued. "The ODP has not disclosed how governance is done, but it is a different model than the preferred and fair meritocracy used by the ASF."

He said questions about the ODP that need answering include:

• Is it pay-to-play? What are the dues and fees?
• How do fees affect voting rights?
• Where do the funds go?
• Why was the ODP announced just as Pivotal exited the Hadoop market and reportedly laid off many engineers?
• Have all of the ODP members really committed? What dues did they pay? Do subsequent members pay the same dues and gain the same voting rights?

"We are committed to community involvement and cooperation to drive technology advancement that drives customer value," Schroeder concluded. "Until it is clear that the structure, focus and participation of the ODP can effectively address key customer concerns, we will continue to focus our efforts on the ASF.

Some of the same concerns voiced by MapR's Schroeder were also presented by Cloudera shortly after the ODP was formed.

"Pivotal and Hortonworks claim that the ODP is driven by an industry-wide longing for standardization in the Apache Hadoop ecosystem," said Mike Olson in a Cloudera blog post. "I don't believe them."