AWS re:Invent 2019: Insert, upsert, and delete data in Amazon S3 using Amazon EMR (ANT239)

[@KannanKalidasan]

Well-designed architecture bringing the OLTP system Commit / Transactions watermarks into file-based distributed systems.

[@kislayapankhuri]

Really well explained!

[@aryankyler7453]

InstaBlaster...

[@srinivasaraghu8574]

Good lecture!

[@adarshverma5429]

But my job has failed due to commit time by throwing an error as "Failed to upsert for commit time ......" What's the reason for this error?

[@awssupport]

I'm so sorry to hear of this trouble, Adarsh. 😔 Check out these docs for troubleshooting steps: go.aws/3Zj9qsT. ⚒️ If you continue to face issues, feel free to engage with the community on our re:Post forum here: go.aws/aws-repost. ^KS.

[@michaelpelts980]

kzbin.info/www/bejne/lZTOf6yClLOlZ6M Isn't MoR better suited for write- or change-heavy workloads with fewer reads? I'm asking because the slide says it's read optimized.