Free tools Windows power users keep installed
One-click scans. No signup required.
“Stop using databases” is a deliberately provocative way to say: stop making every application team run its own database infrastructure when a managed data service can meet the workload’s needs. The application still depends on a database and stored data; the proposed change is who operates the data layer and how developers access it.
What does “use data services” mean?
In a 2021 interview, Patrick McFadin argued for consuming database capabilities through a hosted service or an API rather than taking on the full work of running a database cluster. Instead of configuring infrastructure and database drivers, an application can call a service interface to read and write data. The database technology remains underneath that interface.
As an Amazon Associate I earn from qualifying purchases.
McFadin described the goal as letting important technology “fade into the background” so it is something developers use rather than continually reason through. That is his philosophy, not a measured finding or a universal rule. The phrase does not mean that data modeling, access patterns, or the service’s constraints can be ignored.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →What work might a service take off the application team?
Operating a distributed database can involve provisioning and patching infrastructure, coordinating nodes, configuring drivers, and responding to failures. The 2021 conversation used Cassandra to discuss that burden, including sharding, data models, distributed coordination, and specialist knowledge. These are examples of work teams may need to handle in a self-managed deployment; the exact division of responsibility varies by service and contract.
#1 Best Overall
A managed provider takes responsibility for some infrastructure operations and offers a service boundary—often an API or protocol—for applications. This can reduce operational work, but it does not make the provider responsible for every application-level decision. Teams still need to select appropriate data structures and queries, set permissions, and plan for recovery and cost.
How do self-managed databases and data services differ?
| Decision area | Self-managed database | Managed data service |
|---|---|---|
| Infrastructure operations | Your organization operates and patches the database infrastructure. | The provider operates at least some of the infrastructure; confirm exactly which tasks remain yours. |
| Application access | Your application typically connects using the database’s drivers or protocols. | The service may expose database protocols, APIs, or an access layer; verify supported interfaces. |
| Data model and queries | Your team chooses and works within the database’s model and query capabilities. | Your team still has to check that the service supports the required model and access patterns. |
| Recovery | Your team is responsible for configuring and testing backups and restores. | Available backup and restore controls depend on the service; establish who initiates and verifies recovery. |
| Portability | Migration is still work, but your team controls more of the deployment. | Export options, interfaces, and provider-specific behavior can affect migration effort; assess them before committing. |
| Cost | Account for infrastructure and the people and time needed to operate it. | Account for service charges and usage patterns, as well as any operational work that remains in-house. |
There are no comparable benchmark figures in the cited 2021 sources for cost, performance, staffing, or availability. Compare options using your workload and a realistic recovery and migration scenario instead of assuming that either model is automatically cheaper or faster.
Rank #2
How should you evaluate a data service?
- Describe the workload. List how the application reads and writes data, its data model and query patterns, and its latency and availability requirements.
- Draw the operational boundary. Ask who provisions and patches the system, handles incidents, manages upgrades, and maintains the application’s database integration.
- Walk through recovery. Determine what backups and restore controls are available, who starts a restore, and how your team will test that restored data is usable.
- Review access and security. Check how identities, permissions, and application access are managed for the service you are considering.
- Test portability. Find out how data can be exported and what application changes would be needed to move to another service or deployment model.
- Estimate the full cost. Use expected usage and include the operational work your team will still perform. No price comparison is established by the 2021 interview.
What did the 2021 examples establish—and what did they not?
The Linux Foundation’s exact-title webinar was recorded on March 17, 2021, and sponsored by DataStax. Its description framed cloud and Kubernetes as changing architecture decisions and listed development speed, open source, community, and future-proofing as discussion topics. Those were the event’s planned themes, not independently measured outcomes.
In the June 7, 2021 Serverless Chats interview, McFadin discussed Cassandra, DataStax Astra, and Stargate as examples of hosted database technology and an access layer. They illustrate the argument at that time; they do not establish any product’s current status, feature set, pricing, guarantees, or portability. Check current vendor documentation and terms before relying on a product-specific capability.
McFadin also said, “the era of running your own database is coming to an end. You should not run your own database.” Read that as his strong 2021 opinion, not an established industry-wide conclusion. Self-managed infrastructure can still make sense when a team needs control, has the expertise to operate it, or finds that available services do not fit its workload or constraints.
Quick Recap
Rank #4
- HP ProLiant DL360 G7 8B Server
- 2x X5650 2.66GHz 12-Cores Total
- 32GB RAM / 8x 146GB 10K 2.5in SAS Hard Drives
- P410 w/ 512MB
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
