~/talk

Demystifying Stream Processing with Apache Kafka

Talkat Philly ETE 2016 · on Monday, 2016-04-11· [data-engineering, distributed-systems]
links:none on file
slides:none on file

Argues that Kafka plus an embedded library changes stream processing tradeoffs: Kafka Streams is a lightweight, embeddable API (processor API + DSL) that uses Kafka topics and consumer groups for parallelism, local state backed by RocksDB plus a Kafka changelog for fault recovery, and supports event-time windowing and reprocessing via offsets and log compaction. Claims this reduces Lambda architecture complexity, fits existing deployment tooling, and notes at scale LinkedIn processes ~1.2 trillion messages per day using the same log-based platform.

AI-written summary · about

  1. 0:00Intro
  2. 7:36Myths about Stream Processing
  3. 9:47Summary of Stream Processing
  4. 11:51Stream Processing Challenges
  5. 17:52Lambda Architecture
  6. 22:32What is Kafka
  7. 24:38Partitioning messages
  8. 32:13Batch vs Stream processing
  9. 33:46Stream processing as a consumer
  10. 34:38Tapas Trains
  11. 35:24Library vs Framework
  12. 37:45Kafka Processor API
  13. 40:11Kafka Application
  14. 41:06Execution Model
  15. 43:17Load Balancing
  16. 44:17State Storage
  17. 45:40Map to Physical Processes
  18. 46:40Time
  19. 50:53Kafka Connect
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.