Go Hybrid & Multi-Cloud or Don’t Go
Over the past few months industry analysts have been making some pretty controversial recommendations for data management in the cloud. For a thoughtful and entertaining analysis, I strongly recommend you spend a few minutes watching the keynote session by Pat Moorhead, CEO Moor Insights & Strategy, at the Evolve 2022 Data event in New York. His takeaway: “The world is very much going to be hybrid and multi-cloud.” His delivery—priceless.
Pat isn’t the only analyst talking hybrid and multi-cloud for data management, although he may be the most entertaining. ZDNet had a thorough piece on why single clouds are passé. Workload portability with no cloud lock-in was cited as critical to digital initiative success. Gartner has also been providing very pointed guidance to clients on the urgency and inevitability of hybrid and multi-cloud for data management. If you haven’t been paying attention, here are the key points from my point of view.
Cloud is hybrid and multi
In March 2022, Gartner published A Strategic Roadmap for Migrating Data Management to the Cloud. The big takeaway for me was the urgency of true hybrid and multi-cloud capabilities. These two instructions really struck me:
“The future data ecosystem should leverage distributed data management components—which may run on multiple clouds and/or on premises—but are treated as a cohesive whole with a high degree of automation. Integration, metadata, and governance capabilities glue the individual components together.”
“Through 2026, 90% of data management tools and platforms that fail to support multi-cloud and hybrid capabilities will be set for decommissioning within three years.”
My translation: Don’t start any data management projects without a hybrid and multi-cloud data management platform.
That’s big. None of the hyperscalers do true hybrid and multi-cloud data management today. Most do uni-cloud, a single cloud that sometimes operates closer to the enterprise. It doesn’t manage or analyze data in the datacenter or on premises, but is designed to move data to their uni-cloud. Neither do the point solution (more on them later). Some do multi-cloud, but not hybrid and not really data management. They are more uni-service. Nothing wrong with uni, but it’s not hybrid.
Data management is hybrid and multi
In September Gartner published Innovation Insight: Data Ecosystems Will Reshape the Data Management Market. In my view they doubled down on the March guidance. The most important guidance from my point of view:
“The effective management of data is essential to every aspect of software, analytics, AI/ML, and every digital use case.”
“Hybrid will never disappear; as data ecosystems become the center of gravity for data management, they will enable the integration of data across a hybrid environment.”
“By 2024, 50% of new system deployments in the cloud will be based on a cohesive cloud data ecosystem rather than on manually integrated point solutions.”
My translation: If you don’t have effective data management, you will fail at AI and digital transformation. Data management must be hybrid, otherwise don’t start. And about point solutions: integration costs and lack of hybrid support make them economically and functionally undesirable.
Netting it all out, my read is that Gartner is “all in” on recommending hybrid and multi-cloud and multi-function as essential for data management platforms. The hyperscalers don’t have solutions today. Nor the point solutions for data management and analytics.
Businesses are going hybrid and multi
We are hearing and seeing the same from leading businesses around the world. Hybrid is the new normal. Hybrid data is the future and the future is now.
- A large European multinational insurer implements a central data science factory that leverages both data center and cloud resources to develop new use cases faster than ever.
- A large African telecommunications organization consistently manages and analyzes data from their businesses across the continent where different countries and regions have very different levels of infrastructure maturity.
- One of the world’s largest banks treats cloud infrastructure as a utility, running their data workloads in the cloud that is the most cost effective at the moment.
Cloudera makes hybrid and multi-cloud easier
I encourage you to read both analyst papers thoroughly and checkout Pat’s Evolve 2022 keynote. They are content and controversy rich. And they confirm to me that the direction on data management is becoming very clear:
- Hybrid & multi-cloud is a given
- Hybrid is forever
- Hybrid and open data offer ultimate flexibility
- Point data solutions are a dead end
- Data management at scale is a pre-req for digital
Everyone needs to learn more about hybrid data. Cloudera makes hybrid data easier. Let’s talk.