Real-time TLC with Elasticsearch

Presented at the NoSQL Roadshow SF 2013

Users are not the only life-forms interacting with our systems. Our infrastructure is a complex, evolutionary organism that, like the human body, must be cared for by the information it produces. The ability to respond to health issues is only as effective as the tools we use to capture these data. Elasticsearch is a storage and analytics engine that operates at the scale of reality. The thirty thousand-foot aggregation across billions of metrics is as accessible as the full-text search laser-focused on a single event. Elasticsearch is already used in thousands of enterprises translating data into business decisions.

Transcript

1. Real-Time TLC Drew Raines Elasticsearch
2. Drew
3. Elasticsearch.com
5. Key/value
6. Key/value
7. (Super!)Columns
8. (Super!)Columns
9. Map/Reduce
10. Map/Reduce
11. Schema
12. Schema
14. Real-time
15. High-capacity
16. Dynamic
17. High-capacity Real-time Dynamic
18. Events
20. 2003 Tom moves to San Francisco 2012 Tom moves to NYC
21. name location previous Tom NYC SF
22. name location moved Tom SF 2003 Tom NYC 2012
24. 2003 Tom moves to San Francisco 2009 Betty moves to NYC 2012 Tom marries Betty 2012 Tom moves to NYC
26. High-capacity Real-time Dynamic
27. elasticsearch
28. #hdrdata
29. persistent persistence
30. schema-less -ish
31. fast
32. fast
34. Metrics
35. github.com/etsy/statsd ganglia.sourceforge.net graphite.wikidot.com
39. % git clone git://github.com/elasticsearch/kibana.git % cd kibana % npm install -g http-server \ # or python -m SimpleHTTPServer % http-server Starting up http-server, serving ./ on port: 8080 Hit CTRL-C to stop the server % open http://localhost:8080
40. bin/elasticsearch -f
41. @drewr elasticsearch.com elasticsearch.org github.com/elasticsearch three.kibana.org demo.kibana.org