Part 1 of a 2-part series. Read Part 2.
At SAPPHIRE NOW, I had the chance to sit down with Ken Tsai, global VP and head of Product Marketing, Database and Data Management at SAP, to dive into how he sees the cloud and the architectural flexibility of new data management solutions benefiting companies.
It was interesting speaking with customers at SAPPHIRE NOW and hearing the variety of perspectives on the progress they’d made on their journey to the cloud. From your experience, what sort of use cases lend themselves best to the cloud, and what particular challenges are customers encountering?
Looking at it from the perspective of a holistic integrated data-management solution, it makes sense that when we have data that’s originated in the cloud, we should keep it and manage it directly in the cloud. But this presents some challenges that the technology vendors need to solve – like how to manage data across multiple clouds.
Speaking with customers, we found they don’t just have one cloud; they have multiple clouds, and this cloud data is coming from multiple places (like SaaS solutions, for example, or a CRM service). So the reality is that customer’s raw data is spread across multiple clouds these days.
The promise of a consolidated logical view of the data element is important, but it can’t be achieved by forcing all data into one cloud. We know that’s not feasible. We learned that lesson 20 years ago, when everybody in the world was trying to consolidate data into one single data warehouse! Today management of data across multiple cloud properties is about orchestrating data movement, in a process-flow fashion from one cloud to the other, refining and enriching it (the data), and tracking its lineage and usage. So when you consume the data, you actually have assurance of where that data came from, and that refinement process has already happened. This lets you confidently take action based on the data.
How can a cloud or hybrid (cloud-plus-on-premises) environment benefit companies with traditional on-premises environments?
One thing I want to recommend to all SAP or non-SAP customers is to actually look beyond the deployment options. Certainly putting software in the cloud simplifies management of the software lifecycle, and makes it readily available and more elastic – that’s a given, it’s table stakes. And this is true whether it’s a database platform-as-a-service or software-as-a-service application.
We should also look at the growing accessibility of different operations and services that are already delivered in the cloud through either APIs or deeper functional integrations. Today, think about the possibility of creating an integrated data-processing system, and let’s say it was a best-of-breed machine learning framework, like Google Tensor Flow. Directly running on one of the most advanced infrastructure-as-a-service (IaaS) options and optimized to run on CPU, GPU, or even TPU, you have freedom and flexibility.
The platform would be a collection of technology solutions all operating in the cloud, but that piece of different functionality would be completely dependent on customer choice. It’s up to vendors like SAP to make sure that our different technologies work seamlessly with third-party and open source technologies already running the cloud. Because ultimately, meeting the customers’ needs will drive innovation across the board.
In Part 2 of this series, we’ll examine the advantages of a hybrid approach and how the cloud can help accelerate innovation.