Preventing Data Loss

Ensuring the complete integrity of data for most monolithic applications is a static and manageable concept. Most underlying data stores are atomic, consistent, isolated, and durable (ACID). Once data has been received in the form of a request, a single compiled application will process it through to completion. Threats to the integrity of this operation are few and, more importantly, highly observable and consistent.

For example, a failure in the data store may consistently fail the whole application—or part of the application—until fixed. A failure of code presents itself as a bug that will fail the same way repeatedly for the same given request. Data corruption and loss in monoliths tend to be an all-or-nothing game. If data is being lost, we’ll know through consistent failures to operate.