Building a Cloud Culture at Yelp (BDT305) | AWS re:Invent 2013

1,072 views

Published on

Yelp is evolving from a purely hosted infrastructure environment to running many systems in AWS—paving the way for their growth to 108 million monthly visitors (source: Google Analytics). Embracing a cloud culture reduced reliability issues, sped up the pace of innovation, and helped them support dozens of data-intensive Yelp features, including search relevance, usage graphs, review highlights, spam filtering, and advertising optimizations. Today, Yelp runs 7+ TB hosted databases, 250+ GB compressed logs per day in Amazon S3, and hundreds of Amazon Elastic MapReduce jobs per day. In this session, Yelp engineers share the secrets of their success and show how they achieved big wins with Amazon EMR and open source libraries, policies around development, privacy, and testing.

Published in: Technology, Business
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,072
On SlideShare
0
From Embeds
0
Number of Embeds
9
Actions
Shares
0
Downloads
34
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Building a Cloud Culture at Yelp (BDT305) | AWS re:Invent 2013

  1. 1. Building a Cloud Culture at Yelp Jim Blomo – Engineering Manager, Yelp November 15, 2013 © 2013 Amazon.com, Inc. and its affiliates. All rights reserved. May not be copied, modified, or distributed in whole or in part without the express consent of Amazon.com, Inc.
  2. 2. Why Cloud?
  3. 3. Yelp!
  4. 4. Yelp Data
  5. 5. Why Cloud? specificity & optimization generality & abstraction
  6. 6. Make the Trade-Off!
  7. 7. How Cloud?
  8. 8. Back to the Past
  9. 9. Logging – an aside {user_id: 5, request_id: "b1946ac92492d2347c6235b4d2611184", search_query: "farm to table", city: "Portland, OR", results: [17289, 8230452, 825429, 184312,...], timestamp: 1384469660 }
  10. 10. Hadoop Trade-Offs
  11. 11. Riddle: Success? • How do you know when your infrastructure is a success? • Starts failing under heavy load • "Too many" users overloading the system
  12. 12. Hadoop Issues
  13. 13. EMR Solutions S3 • Clusters up for limited amount of time • Upgrades handled by Amazon • Multiple clusters means no capacity coordination
  14. 14. Trade-Offs
  15. 15. Trade-Offs
  16. 16. Trade-Offs Standard configs Resource consumption tracking testing
  17. 17. No Cargo Cults
  18. 18. Standard Configs
  19. 19. mrjob Configs # standard # memory intensive # cpu intensive
  20. 20. Resource Tracking python -m mrjob.tools.emr.terminate_idle_job_flows -c mrjob.conf
  21. 21. Testing --runner emr mr_canary.py
  22. 22. mrjob is Open Source
  23. 23. Adoption
  24. 24. Cloud Calculations • 5 days with 10 machines = 1 day with 50 machines • (On demand pricing simplification)
  25. 25. Cost
  26. 26. Cost
  27. 27. Cost Control
  28. 28. Data Availability • s3mysqldump
  29. 29. Overview in: logs & DB dumps in: JSON logs out: CSV, JSON, MyISAM in: s3mysqldump S3 out: LOAD DATA or rsync
  30. 30. Pitfalls
  31. 31. Leaky Abstractions
  32. 32. Closed Source bootstrap-actions/configure-hadoop mapred.reduce.tasks.speculative.execution=false
  33. 33. Data Explosion Tron
  34. 34. Wanted
  35. 35. Next Up: Services
  36. 36. Cloud Strategy Target generality & abstraction easiest way
  37. 37. Tandem
  38. 38. We’re Hiring yelp.com/careers
  39. 39. Please give us your feedback on this presentation BDT305 – Cloud @Yelp As a thank you, we will select prize winners daily for completed surveys!

×