AWS re:Invent 2018: [REPEAT 1] Databases on AWS: The Right Tool for the Right Job (DAT205-R1)

[@ruixue6955]

19:40 customer example: Airbnb 20:08 *DynamoDB* , key-value, for *search user history* 20:15 *ElastiCache* for *Session State* 20:23 Amazon *RDS* for *relational data* 20:38 Duolingo 21:18 *DynamoDB* for item tracking 21:23 *Aurora* - primary transactional database for *user data* 21:29 *ElasticCache* - caching layer to speed up descriptions on learnings around keywords such as "da", "and", "it"

[@ruixue6955]

8:41 *common data categories and use cases* 10:06 relational, key-value, graph 11:03 relational 12:13 use case - doctors affiliated with Mercy hospital 12:53 use case - number of patient visits each doctor completed last week 13:43 key-value 14:55 use case - Status of Hammer57 15:45 use case - Return all data associated with Hammer57 16:19 graph 17:08 use case - product recommendation to a user 18:49 use case - identify a common friend and make a recommendation

[@ruixue6955]

1:24 what is your database strategy 2:27 Lift and shift 4:30 Aurora 5:09 RDS (Amazon) Relational Database Service 5:47 DMS (AWS) Data Migration Service

[@ruixue6955]

33:50 Ledger database 34:20 39:00 fundamental key concept of ledger database 47:40 time series 47:45 what is time-series data example: 47:56 weather 48:00 stock price 48:45 what is special about time-series database

[@Nil-js4bf]

Love the bookstore demo showcasing how a site's features are split into different "services" and how different DB types are picked for each feature.

[@ruixue6955]

27:52 27:57 why *Redis* 29:50 recommendation engine 30:59 when I click on this book, I also want it to present other books that people has purchased that also bought this book 32:39 summary

[@TusharKale9]

Very well explained different AWS databases suite, Shawn. Thank you, Tushar

[@viditshah8840]

THIS IS GOLD STANDARD

[@aprilzhang8856]

I learnt so much. Great talk!! Thank you!!

[@daytraderph]

great talk

[@austinmw89]

What would be the most performant way to store and perform analytics on image metadata from neural network object detection and instance segmentation results?

[@ruixue6955]

53:45 summarize

[@TrippleXD545]

it feels like every third thing you say is bs

[@TrippleXD545]

also. i don't think you met those people "yesterday", seeing as your presentation is rather dependent on those slides -- it would be awfully convenient