Raffi Krikorian, Twitter Timelines at Scale
•
15 gefällt mir•17,700 views
Charla de Raffi Krikorian, el VP de Ingeniería de Twitter, sobre la infraestructura de los Timelines de twitter
Empfohlen
Empfohlen
Weitere ähnliche Inhalte
Was ist angesagt?
Was ist angesagt? (20)
Andere mochten auch
Andere mochten auch (20)
Ähnlich wie Raffi Krikorian, Twitter Timelines at Scale
Ähnlich wie Raffi Krikorian, Twitter Timelines at Scale (20)
Mehr von Mariano Amartino
Mehr von Mariano Amartino (20)
Kürzlich hochgeladen
Kürzlich hochgeladen (20)
Raffi Krikorian, Twitter Timelines at Scale
- 1. timelines at scale @raffi qcon sf 2012
- 4. Pull Push Targeted twitter.com home_timeline API User / Site Streams Mobile Push (SMS, etc.) Queried Search API Track / Follow Streams
- 5. the challenge ⇢> 150M world wide active users ⇢> 300K QPS for timelines ⇢naïve timeline “materialization” can be slow
- 8. Timeline Service Ingester SearchCache Redis Redis Earlybird Blender PushCompute HTTP Push Mobile Push BatchCompute Hadoop Write API Fanout Redis Redis TimelineCache Redis Social Graph Service insert ⇢keyed off “recipient” ⇢pipelined 4k “destinations” at a time ⇢replicated
- 9. Timeline Service Ingester SearchCache Redis Redis Earlybird Blender PushCompute HTTP Push Mobile Push BatchCompute Hadoop Write API Fanout Redis Redis TimelineCache Redis using redis ⇢native list structure Tweet ID BitsUser ID 8 bytes 4 bytes8 bytes
- 10. Timeline Service Ingester SearchCache Redis Redis Earlybird Blender PushCompute HTTP Push Mobile Push BatchCompute Hadoop Write API Fanout Redis Redis TimelineCache Redis using redis ⇢native list structure ⇢RPUSHX to only add to cached timelines Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID BitsUser ID Tweet ID Tweet ID Tweet ID
- 13. Pull Push Targeted twitter.com home_timeline API User / Site Streams Mobile Push (SMS, etc.) Queried Search API Track / Follow Streams
- 15. PushCompute HTTP Push Mobile Push BatchCompute HadoopSearchIndex Blender Redis Timeline Service Ingester Earlybird Write API Fanout Redis Redis TimelineCache RedisEarlybird blender ⇢queries one replica of all indexes ⇢merges & ranks results
- 18. the challenge (part #2) ⇢fanout can be really slow! ⇢...especially for high follower counts
- 20. @barackobama 23 million followers 31 million followers @katyperry 28 million followers @justinbieber 28 million followers @raffi 0.019 million followers @ladygaga
- 21. there are over 400 million tweets a day
- 22. a second 4600 tweets 0.2 m a twe ≈
- 24. Write API Ingester Fanout SearchIndexRedis Earlybird Earlybird Redis Redis Redis TimelineCache search index ⇢[‘hello’,‘world’] fanout index ⇢[@danadanger, ...]
- 25. User Intent Query Expansion “Hello, world” “Hello” AND “world” @raffi’s home timeline home_timeline:raffi
- 26. User Intent Query Expansion “Hello, world” “Hello” AND “world” @raffi’s home timeline user_timeline:nelson OR user_timeline:danadanger
- 27. User Intent Query Expansion “Hello, world” “Hello” AND “world” @raffi’s home timeline home_timeline:raffi
- 28. User Intent Query Expansion “Hello, world” “Hello” AND “world” @raffi’s home timeline home_timeline:raffi OR user_timeline:taylorswift13
- 30. Asynchronous Path Query Path BatchCompute Hadoop Synchronous Path PushCompute HTTP Push SearchIndex Blender Redis Timeline Service Ingester Earlybird Write API Fanout Redis Redis TimelineCache RedisEarlybird Mobile Push
- 31. Synchronous Path Query Path BatchCompute Hadoop Asynchronous Path PushCompute HTTP Push SearchIndex Blender Redis Timeline Service Ingester Earlybird Write API Fanout Redis Redis TimelineCache RedisEarlybird Mobile Push
- 32. Asynchronous Path Synchronous Path BatchCompute Hadoop Query Path PushCompute HTTP Push SearchIndex Blender Redis Timeline Service Ingester Earlybird Write API Fanout Redis Redis TimelineCache RedisEarlybird Mobile Push
- 33. timeline query statistics ⇢>150m active users worldwide ⇢>300k qps poll-based timelines @ 1ms p50 / 4ms p99 ⇢>30k qps search-based timelines
- 34. tweet input ⇢~400m tweets per day ⇢~5K/sec daily average ⇢~7K/sec daily peak ⇢>12K/sec during large events
- 35. timeline delivery statistics ⇢30b deliveries / day (~21m / min) ⇢3.5 seconds @ p50 to deliver to 1m ⇢~300k deliveries / sec
- 42. thanks!