Virtualization and Cloud executives share their predictions for 2017. Read them in this 9th annual VMblog.com series exclusive.
Contributed by Roman Shaposhnik, Vice President of Technology for ODPi
Looking Forward: 2017 Predictions for the Rapidly-Growing and Ever-Evolving Big Data Landscape
ODPi is a nonprofit organization committed to simplification & standardization of the big data ecosystem. As the VP of Technology for the Linux Foundation project and PMC member of Apache Bigtop, Groovy, Geode, Ignite and Incubator; Committer to Apache Hadoop and Giraph; and Mentor for Apache HAWQ (incubating), Apache MADlib (incubating) and Apache Fineract (incubating) at Apache Software Foundation, I'm well-versed in Open Source trend identification and executing on strategies that complement these developments for the benefit of our community. I've put my trend observations for the big data landscape into the below predictions for the coming year.
- 2017 will see a huge rise in SI (System Integration) company launches, driving a lot of value and technological innovation. Big data has officially graduated to a plateau of productivity and SI companies are exactly the type of business to drive further adoption in this space. After 2016, the year of whizz-bang features in big data platforms is over. From 2017 on, it will all be about fit-n-finish capabilities.
- In the coming year, cloud and on-premises analytics are going to diverge even further than they did in 2016. As always, Amazon showed everyone this year where the public cloud is headed when it comes to deriving business value from an application's data exhaust. If you're an Amazon customer, all your data exhaust is in S3 anyway and, thanks to Amazon Athena, you can now do seamless analytics without any kind of ETL.
- With open source cloud platforms, like Cloud Foundry, Kubernetes, Docker and more, driving de-facto standardization in the application cloud management space, I believe that the Hadoop ecosystem will finally be boarded on the heterogeneous cloud in a portable way. Given that both Kubernetes and Hadoop have Linux Foundation efforts helping their enterprise adoption, I expect the interplay between the two to play a significant role in 2017.
- In the next year, the Hadoop ecosystem will finally wise up to hybrid data warehousing architectures - allowing open source MPP (Massively Parallel Processing) technologies, like Greenplum, to assume their rightful place in the big data architecture bingo score card.
- I see the growing confluence of 12-factor application design and ease of integrating meaningful analytics with application data exhaust will finally result in explosive innovation - especially in the smart, data-driven app space - during 2017. This will further remove human involvement from the interactions between businesses and their customers.
##
About the Author
Roman Shaposhnik is a well-known thought leader and innovator in the field of open source enterprise software development. Currently he is a VP of Technology for ODPi at Linux Foundation and Director of Open Source strategy at Pivotal Inc. In the open source community he is known as one of the top contributors to the Apache Software Foundation (committer on Apache Hadoop and founder of Apache Bigtop) and a former VP of Apache Incubator. Roman has been involved in hands-on Open Source software development for more than two decades and has hacked projects ranging from Linux kernel to the flagship multimedia library FFmpeg. Ultimately, though, his core identity is still that of a member of lost tribes of Sun Microsystems still wandering in the valley.