Eventbrite dataplatform and services - Interest graph based recommendations
Download as PPTX, PDF3 likes3,265 views
The document discusses Eventbrite's data platform and services. It describes Eventbrite as a social event ticketing platform that has doubled revenue yearly. It then summarizes Eventbrite's use of Hadoop, HBase, MongoDB and other technologies to power analytics, recommendations, and real-time data processing. Recommendations are generated using interest graphs and social graphs built from user activity data stored in Hadoop and HBase.
![Feature Generation
• Mixed Features
• A series of map-reduce jobs
• Output on HDFS in flat files; Input to subsequent jobs
• Orders = Event Attendees
• MAP: eid: uid
• REDUCE: eid:[uid]
• Attendees Social Graph
• Input: eid:[uid]
• MAP: uidi:[uid]
• REDUCE: uid:[neighbors]
• Interest based features, user specific, graph mining etc
• Upload feature values to HBase](https://image.slidesharecdn.com/strata-2012-120302174911-phpapp01/85/Eventbrite-dataplatform-and-services-Interest-graph-based-recommendations-25-320.jpg)
Eventbrite dataplatform and services - Interest graph based recommendations
- 1. Data Platform and Services Vipul Sharma and EyalReuveni
- 2. Agenda Eventbrite Data Products Data Platform Recommendations Questions
- 3. • A social event ticketing and discovery platform • 50th Million Ticket Sold • Revenue doubled YOY • 180 Employees in SOMA SF • Solving significant engineering problems • Data • Data, Infrastructure, Mobile, Web, Scale, Ops, QA • Firing all cylinders and hiring blazing fast www.eventbrite.com/jobs
- 7. Analytics • Add–Hoc queries by Analysts
- 11. Hadoop Cluster • 30 persistent EC2 High-Memory Instances • 30TB disk with replication factor of 2, ext3 formatted • CDH3 • Fair Scheduler • HBase
- 12. Infrastructure • Search • Solr • Incremental updates towards event driven • Recommendation/Graph • Hadoop • Native Java MapReduce • Bash for workflow • Persistence • MySql • HDFS • HBase • MongoDB (Investigating Cassandra and Riak)
- 13. Infrastructure • Stream • RabbitMQ • Internal Fire hose (Investigating Kafka) • Offline • MapRedude • Streaming • Hive • Hue
- 14. Infrastructure - Sqoozie • Workflow for mysql imports to HDFS • Generate Sqoop commands • Run these imports in parallel • Transparent to schema changes • Include or exclude on column, data types, table level • Data Type Casting tinyint(1) Integer • Distributed Table Imports
- 15. Infrastructure - Blammo • Raw logs are imported to HDFS via flume • Almost real-time – 5 min latency • Logs are key-value pairs in JSON • Each log producer publishes schema in yaml • Hive schema and schema yaml in sync using thrift • Control exclusion and inclusion
- 16. Recommendations
- 17. You will like to attend this event
- 18. Recommendation Engines Interest Graph Based Social Graph Based (Your (Your friends who friends like Lady like rock music Collaborative Gaga so you will like you are Filtering – Item- like Lady attending Eric Item similarity Gaga, PYMK – Clapton Event– Facebook, Linkedin Eventbrite) Collaborative (You like Godfather so you ) Filtering – User- User Similarity will like Scarface - Netflix) (People who Item bought camera Hierarchy also bought batteries - (You bought Amazon) camera so you need batteries - Amazon)
- 19. Why Interest? Events are Social Events are Interest Dense Graph is Irrelevant Interest are Changing
- 20. How do we know your Interest? • We ask you • Based on your activity • Events Attended • Events Browsed • Facebook Interests • User Interest has to match Event category • Static • Machine Learning • Logistic Regression using MLE • Sparse Matrix is generated using MapReduce • A model for each interest
- 21. Model Based vs Clustering Item-Item vs User-User Building Social Graph is Clustering Step Social Graph Recommendation is a Ranking Problem
- 22. Implicit Social Graph U1 E1 E4 U2 U3 E2 E3 U4 U5
- 23. Mixed Social Graph U1 E1 U2 U3 E2 E3 FB U4 U5 LI
- 24. 15M * 260 * 260 = 1.14 Trillion Edges 4Billion edges ranked Each node is a feature vector representing a User Each edge is a feature vector representing a Relationship
- 25. Feature Generation • Mixed Features • A series of map-reduce jobs • Output on HDFS in flat files; Input to subsequent jobs • Orders = Event Attendees • MAP: eid: uid • REDUCE: eid:[uid] • Attendees Social Graph • Input: eid:[uid] • MAP: uidi:[uid] • REDUCE: uid:[neighbors] • Interest based features, user specific, graph mining etc • Upload feature values to HBase
- 26. U1 U2 U3
- 27. HBase
- 28. HBase • Collect data from multiple Map Reduce jobs • Stores entire social graph • Over one million writes per second
- 29. HBase rowid neighbors events featureX 2718282 101 3 0.3678795
- 30. HBase rowid 314159:n 314159:e 314159:fx 161803:n 161803:e 161803:fx 2718282 31 1 0.3183 83 2 0.618
- 31. Tips & Tricks • Distributed cache database • Sped up some Map Reduce jobs by hours • Be sure to use counters!
- 32. Tips & Tricks • Hive (ab)uses • Almost as many hive jobs as custom ones • “flip join” • Statistical functions using hive • UDF
- 33. Tips & Tricks • Memory Memory Memory • LZO, WAL • Combiners are great until • Shuffle and Sorting stage • Hadoop ecosystem is still new
- 34. Questions?