Last week was interesting. I spent most of it in Las Vegas at EMC World, which was as much about Dell as it was about EMC itself. There was a ceremonial handoff from Joe Tucci to Michael Dell to lead the new combined entity, but Jeremy Burton was perhaps even more in the spotlight as he outlined the vision.
Much of this vision was about the balance between traditional data center environments and something he called "cloud-native" applications. Substituting "next-gen" for "cloud-native" might be more accurate, as this category included everything from PaaS to big data to containers to hybrid clouds. Hadoop, Cassandra, and MongoDB were cited as examples as cloud-native, which felt odd. Certainly they are cloud-friendly, but they're by no means exclusively cloud-centric. See my last post on cloud big data for more thoughts on this topic.
Indeed, much more of the emphasis on big data environments was on data lakes on-premises. EMC Isilon is being held forth as the preferred multi-protocol, scale-out platform here. For more intensive and more active big data workloads, (including streaming) EMC pointed toward the promising DSSD all-flash storage. Going a step further, for a ready-to-rock big data appliance, VCE was mentioned, even as there were heavy hints that Dell servers would help underpin a new generation of engineered systems.
What was startlingly missing was the software portfolio around database management, big data, and analytics. Pivotal Big Data Suite was originally born from a combination of tools, including Greenplum, Gemfire, Hawq, and Pivotal Hadoop. Putting this together with Dell's software offerings like TOAD, Boomi, and Statistica would make a lot of sense. It doesn't appear that this will happen.
I asked Michael Dell and Joe Tucci about their plan on the software side, and the answers were not at all reassuring. The use cases for Pivotal were all around Cloud Foundry and Pivotal Labs development. I know both Dell and Pivotal are successful and growing in their data management software businesses, but I think they are overlooking an opportunity to build something even bigger in the big data space with all the assets together. Instead, the focus here is clearly on the hardware and cloud facilitation.
Speaking of Pivotal assets, I'm pleased to see spin-out Snappy Data has raised an A round for their Spark-based, in-memory hybrid transaction and analytics database.
Which brings me to the biggest surprise of the week — the announcement of $253 million in a new round for Pivotal, including Microsoft and the Ford Motor Company. As Microsoft has already welcomed Hortonworks and Cloudera to its Azure big data offerings, there didn't seem to be much need for another data platform there, other than perhaps for customer choice. The huge investment by Ford can only be about building smart cars, so this seems to play toward faster development of autonomous driving and advanced IoT applications, and big data will no doubt be a key part of these solutions. Interesting times indeed.