Streaming Data
Definition
A data stream continuously delivers time-related events that can be processed incrementally.
A data stream is a potentially unlimited sequence of time-related data events that continuously arrive and are often processed incrementally. Event time and arrival order may differ.
Simply explained
Sensors continuously send new messages. Some arrive late, so their arrival order does not necessarily correspond to the order of events.
Why it matters in practice
The term improves reliable access to data and knowledge as well as their usable linkage in digital systems.
Example from business
Machine sensors send events that are aggregated in time windows and checked for anomalies.
Limits and pitfalls
- Assuming a complete global order, mixing event time and processing time, or handling late, duplicate, and out-of-order events without watermark, idempotency, and retry rules.
Origin & history
Stream processing evolved from event processing, databases, and messaging systems into a central pattern of modern data architectures.
Use in mid-sized companies
- Possible application: Machine sensors send events that are aggregated in time windows and checked for anomalies.
Frequently asked questions
What does data streams mean?
A data stream is a potentially unlimited sequence of time-related data events that continuously arrive and are often processed incrementally. Event time and arrival order can differ. What mistakes should be avoided? Assuming a complete global order, mixing event time and processing time, or handling late, duplicate, and out-of-order events without watermark, idempotency, and retry rules.
Sources
- The Dataflow Model · Akidau et al., The Dataflow Model
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: