~/talk

RBEA- Scalable Real-Time Analytics at King

Talkat Flink Forward Berlin 2016 · 2016-09-12..09-14· [data-engineering]
github:gyfora
slides:none on file

RBEA (Rule-Based Event Aggregator) is a Groovy DSL platform that compiles user scripts into one heavy Apache Flink job per game, injects scripts into the running operator, and routes generated output events to sinks like Kafka, MySQL or HDFS. It keeps per-user state in a serialized value state on RocksDB state backend with a custom cache, uses a dynamic window assigner to support arbitrary window sizes without proliferating operators, and handles scaling for King’s ~400M users and ~30 billion events/day — it works in production and speeds analyst iteration.

AI-written summary · about

  1. 0:00Introduction to RBEA
  2. 1:52About King and its data
  3. 6:09RBEA platform goals
  4. 8:39RBEA architecture overview
  5. 13:13RBEA Flink backend details
  6. 15:43RBEA API and scripting
  7. 24:19Aggregation and output handling
  8. 26:29User state management
  9. 29:00RBEA usability benefits
  10. 31:30Summary and Q&A
~/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.