DataOps is still a relatively new concept, it combines people, processes, and technology to build and enhance data analytics.
At Data Summit 2019 Kevin Petrie, Senior Director of Marketing, Attunity, Inc., explored the the five key steps necessary to be successful with DataOps, including the process and cultural shift required during his session, “Five Key Requirements for DataOps Success.”
“We’re a software vendor so we don’t have a magical sword or magical tool that will enable DataOps success but with that said it does involve some tools and technology,” Petrie said. “DataOps is an emerging discipline that helps build and manage more efficient, more effective data pipelines.”
The world has changed very quickly, with moves to multi/hybrid cloud environments and the need to replicate data across many different platforms, Petrie explained.
“People are simply moving data across too many different components,” Petrie said.
Data consumers are driving a lot of demands and as a result projects are failing due to this friction, according to Petrie.
The first steps to successful DataOps should build on continuous data integration and testing. Because communication gaps between data managers and consumers persist, Attunity created a platform that will address this.
The five steps include:
- Real time data delivery
- Data lake automation
- Data warehouse automation
- Faster development lifecycle
- Metadata and control
Bonus requirement includes fostering a new culture at the organization that ensures teams are collaborating effectively, Petrie said.
“Find a project that broke and failed and then find someone who’s good at cross functional collaboration,” Petrie said.
Many Data Summit 2019 presentations are available for review at https://www.dbta.com/DataSummit/2019/Presentations.aspx.