After coming to the meetup event, http://www.grokkingengineering.org, it's very informative event and lots of useful information. Then I have some new ideas , that could solve problems I get during development and deployment using Lambda Architecture in 1 year production at FPT Online, where tons of data hit the server and all system have to make decisions, checking, booking, mining , bidding in near-real-time.
If you miss this talk, you can check this link for more details http://engineering.viki.com/blog/2014/data-warehouse-and-analytics-infrastructure-at-viki/
Some main issues when deploying and implementing the lambda arch in practice:
not Agile enough "The problem with the Lambda Architecture is that maintaining code that needs to produce the same result in two complex distributed systems is exactly as painful as it seems like it would be. I don’t think this problem is fixable."Refer link http://radar.oreilly.com/2014/07/questioning-the-lambda-architecture.html
=> Keep it simple as possible, but not simpler
We are in the dawn of "Fast Data"
https://www.google.com/events/io/schedule/session/793fe228-89c5-e311-b297-00155d5066d7
http://radar.oreilly.com/2014/12/fast-data-calls-for-new-ways-to-manage-its-flow.html
http://www.infoworld.com/article/2608040/big-data/fast-data--the-next-step-after-big-data.html
http://www.wired.com/2013/04/big-data-fast-data-smart-data/
Using User Tracking with
http://en.wikipedia.org/wiki/Real-time_business_intelligence
=> Fast data is really popular in practice, real-time marketing and marketing automation is the new trend
I made simple diagram here, for easy understanding, combines cool ideas from http://www.reactive-streams.org with Lambda Architecture to create a new system that reacts faster with 2 types of new data:
- behavioral data (game, interests, sessions, touch, ...)
- context data (location, time, referrer, lead ,...)
  |
| Apply Marketing Automation with RFX framework |
