Skip to content
dvnt.io

Resources · glossary

Database

The place where a system keeps its data in an organised form: products, orders, accounts, content. Everything else can be rebuilt; the database cannot.

Also called
  • database
  • DBMS
  • database engine
  • SQL database
  1. 01

    Why it is the most important part

    Because code can be written again and a server set up from scratch. Lost data does not come back. That is why backups and the ability to restore the state from a specific hour matter more than the choice of technology.

  2. 02

    Relational or document

    Relational databases (PostgreSQL, MySQL) enforce consistency: an order cannot exist without a customer. Document databases (MongoDB) are more flexible with data whose shape changes. For most web projects both work; the difference shows at scale and in reporting.

  3. 03

    A question worth asking

    Where the database runs and who makes the backups. A database on the same server as the application is cheaper and is enough for a long time, but it can slow down during deployments, because building the new version takes its resources.

  4. 04

    An unrestored backup is a hypothesis

    A dump once a day is enough only if you accept losing a day's work. If not, you need point-in-time recovery, and a rehearsed procedure, not just a file on a disk.

In practice

Related terms

  • APIAn agreed way for one system to ask another for data or tell it to do something. APIs are how a shop knows its stock levels and how a website sends a contact to the CRM.
  • VPS (virtual private server)A dedicated slice of a physical server with guaranteed resources and full control over the configuration. The middle ground between shared hosting and your own machine.

← All terms

15 minutes. No sales pitch.

A term in a proposal sounds foreign? In 15 minutes we explain what it means for your project.