What is Batch Processing

Definition

Batch processing is the execution of data jobs on accumulated groups of records at scheduled intervals rather than continuously, processing large volumes efficiently in a single run when immediate, record-by-record results are not required.

« Back to Glossary Index

Processes very large volumes efficiently by handling records in bulk
Simplifies logic and recovery since each run has a clear start, end, and scope
Uses compute economically by scheduling heavy work during off-peak windows
Fits reporting cycles where periodic, complete results matter more than instant ones

Real World Example

A payroll system runs a nightly batch job that processes all the day's timesheets at once, calculating pay, taxes, and deductions in a single scheduled run rather than recomputing after every individual entry.

FAQs

When is batch processing appropriate?

When results are needed periodically rather than instantly, and processing large volumes together is more efficient than handling each record live.

How does batch differ from stream processing?

Batch processes accumulated data on a schedule, while stream processing handles each event continuously as it arrives.

What are common batch workloads?

Nightly ETL, payroll, billing, report generation, and large-scale model training are typical batch jobs.