iRobot: Serverless Data Cataloging for Data Scientists

2 949
22.9
Опубликовано 6 ноября 2019, 18:47
iRobot's serverless architecture for their connected robot fleet has been well documented in AWS re:Invent sessions. Recently, the iRobot data science team went into production streaming real-time event data from millions of robots into S3 and DynamoDB into a serverless ETL pipeline (SQS, Lambda, Kinesis, Kinesis Firehose) and landing in an S3 bucket, where Glue jobs perform additional transformation. The resulting data is analyzed using Athena and used to populate fleet reports.

Subscribe:
More AWS videos bit.ly/2O3zS75
More AWS events videos bit.ly/316g9t4

#AWS
Свежие видео
3 дня – 132 9872:34
6′46″874 | Trailer
24 дня – 12 8011:41
AI Summit DC 2024 Highlights
автотехномузыкадетское