Mongo: Performance and Troubleshooting

Transcript

1. Performance & Troubleshooting @kirilnyc @gcsports
2. • How MongoDB works (layman's version) • Common failure cases • Best practices
3. Fundamentals • OS Pager, LRU cache ejection • Working Set and implications • Documents on disk
4. Virtual Memory LRU
5. Working Set
6. Documents on Disk
7. Failing • Underestimating Working Set • Ill-Fitting Use Cases • Schema Design Mistakes
8. Oops, Overload
9. Estimating Working Set • Indexes • Core operational data (user records, etc) • Secondary records (logs, sessions) • Long tail data (historical, related) • Scans*
10. I Know, Let's use Mongo!
11. Sub-optimal Use Cases • Session storage • Big fragmented collections • Giant working sets + performance demands • Clearly tabular data
12. Simulated Joins!!!
13. Let's NoSQL! • Look for the largest granularity that works • Eschew lookup collections • Prefer containment over reference • Query sparingly
14. Best Practices • Denormalize heavily • Do Capacity Planning • Live in your slow query logs • Watch the numbers
15. Documents No thanks. Yes please.
16. Capacity Planning
17. Logs *yawn* Ruh-roh...
18. Numbers • Query load (subjective, learn yours) • Lock percentage (< 50%) • Queues (single digits) • Page faults (single digits)
19. @kirilnyc CTO, GameChanger Media http://GC.io/about