Kappa Architecture
A simplification of lambda that handles both real-time and reprocessing with one streaming path.
Kappa architecture is a data-processing design that drops lambda’s separate batch layer and treats everything as a stream. All data flows through one streaming pipeline; when you need to recompute history (a bug fix or new logic), you replay the immutable event log through the same code rather than maintaining a parallel batch job.
Worked example: events land in a durable log like Kafka with long retention; the streaming app computes live results, and to backfill a corrected metric you simply reset the consumer to offset zero and reprocess the whole log through the identical code. Gotcha: it only works cleanly when your source of truth is a replayable, retained event log — if events expire or are mutated in place, you lose the ability to reprocess, which is the whole premise kappa depends on.