For this initial Data Services Manager post on VirtualBlocks, I would like to provide some context around DBaaS, why VMware is helping our customers to implement a DBaaS on vSphere and how it fits into our “Cloud Smart” strategy.
What is DBaaS?
Database as a Service (DBaaS) concept originated as a cloud computing service that provided access to a database without the need to set up physical hardware, install software, or configure the database. Today, all public cloud providers are offering (at least one) such service with the promises of reducing costs, improving scalability and overall simplifying database management. But the reality is more nuanced, and customers are now looking at ways to avoid “Cloud Chaos” to become “Cloud Smart” (hear from VMware CEO Raghu Raghuram here).
Why are customers increasingly considering on-premises DBaaS?
While public cloud-based DBaaS have been and still are very popular, we see an increasing number of customers looking for on-premises DBaaS for multiple reasons such as data privacy, data security, data sovereignty, low latency, high performance, predictable cost.
On-premises DBaaS platform gives enterprises the most flexibility and control over their database environment. More specifically, customers are looking for:
- Control: choose the hardware and software that your database runs on.
- Security: implement your own security policies and procedures.
- Compliance: ensure that your database meets your organization’s compliance requirements.
But building a custom DBaaS on-premises comes with some challenges:
- Cost: upfront costs of setting up and managing a DBaaS platform can be high.
- Complexity: hard to set up and manage a DBaaS platform.
- Skills: expertise to build and manage a DBaaS platform.
The above represents the typical tradeoffs of “Build” Vs. “Buy”.
VMware Data Services Manager
VMware Data services Manager is a software solution that provides the same convenience as a public cloud-based DBaaS platform, but it is installed and managed on-premises eliminating the challenges of building a custom on-premises DBaaS platform. It helps enterprise IT team stays in control, automate day 2 operations to free up valuable DBA time and provide App developers with the agility they need.
You can learn more about Data Services Manager here and test it yourself with the Hands-on Lab here.
What trends do we see?
1- Multi-clouds enabled.
The reality is that most of customers have to manage at least 2 clouds (one of them being often time their existing a private cloud). They have to cope with different types of infrastructure that results in silos. Not only of physical infrastructure but also operational models, security, compliance. Running one’s applications across those silos is complex, expensive and risky. VMware’s strategy is to offer a layer of common abstractions for operations and a common platform for running applications, on any type of cloud. One aspect of those common abstractions have to do with data. Data is at the epicenter of our customers’ digital transformation.
2- Postgres-compatible as building block.
We see Postgres and Postgres-compatible engines lead the relational (and opensource) databases of choice for customers as not only it has high performance, lower cost but it also ensures portability in a multi-cloud world. Of course, with the emergence of new use cases like AI/ML, customers will continue to look for best-of-breed data engines as well but it shouldn’t mean additional operational complexity to manage those diverse data engines. Finally, after the hype, customers are starting to plan more effectively for the cost of leveraging AI to stay ahead of the competition. I would recommend the great A16Z blog post on Navigating the High Cost of AI Compute as a first reference point.
3- Operating models choice.
Finally, customers want choice in their operating models:
- Self-managed: for control, flexibility and cost (post a certain scale), customers are managing their infrastructure.
- Partner-managed: customers, who have to ensure data sovereignty like in Europe, are looking at geo-specific partner(s) to manage some of their infrastructure and data services.
- Vendor-managed: for quick start and convenience, customers are using vendor-managed offerings but like the location of IT, the operational model choice is more nuanced with some tradeoffs to consider.
Conclusion
Data is at the center of the digital transformation and the DBaaS of the future will have an broader scope than the early days by being multi-clouds, multi data engines (beyond databases) and the foundation of the customers data platform. If you want to hear more about VMware Cloud Data infrastructure strategy and portfolio: join us at VMware Explore in Las Vegas (Aug 21-24):
- Meet the Expert Roundtable: Implementing Database as a Service for vSphere (CEIM2546LV)
- Hands-on Labs Expert-led Workshop: VMware Data Services Manager – Getting Started (HOL-2407-01-SDC)
- Breakout Session: Cloud Data Infrastructure: Strategy and Product Portfolio (VIB1643LV)
