hudi VS kudu

Apache Iceberg is one of the three types of lakehouse, the other two are Apache Hudi and Delta Lake.

Structured, Semi-structured and Unstructured can be stored in one single format, a lakehouse storage format like Delta, Iceberg or Hudi (assuming those don't require low-latency SLAs like subsecond).

Apache Hudi

Try Apache Hudi, it is fully integrated with AWS and offers almost everything that you requested.

the corresponding kafka topics have 30d retention and I intend on having s3 sink connector for long term storage (open to other ideas here too, I noticed theres a hudi connector also)

Apache Hudi

You could have a look at Apache Hudi - especially if you're running your Data Pipelines using Spark or Flink.

https://github.com/apache/kudu/blob/master/examples/quickstart/impala/README.adoc https://medium.com/@nifi.notes/building-an-effective-nifi-flow-replacetext-60a6016d378c https://community.cloudera.com/t5/Community-Articles/Running-DNS-and-Domain-Scanning-Tools-From-Apache-NiFi/ta-p/248484 https://community.cloudera.com/t5/Community-Articles/Using-Cloudera-Data-Science-Workbench-with-Apache-NiFi-and/ta-p/249469 https://community.cloudera.com/t5/Community-Articles/Scanning-Documents-into-Data-Lakes-via-Tesseract-MQTT-Python/ta-p/248492 https://community.cloudera.com/t5/Community-Articles/Adding-Stanford-CoreNLP-To-Big-Data-Pipelines-Apache-NiFi-1/ta-p/249378 https://community.cloudera.com/t5/Community-Articles/Using-Apache-NiFi-for-Speech-Processing-Speech-to-Text-with/ta-p/249242 https://community.cloudera.com/t5/Community-Articles/Ingesting-Flight-Data-ADS-B-USB-Receiver-with-Apache-NiFi-1/ta-p/247940 https://community.cloudera.com/t5/Community-Articles/Integrating-lucene-geo-gazetteer-For-Geo-Parsing-with-Apache/ta-p/247993 https://community.cloudera.com/t5/Community-Articles/Creating-WordClouds-From-DataFlows-with-Apache-NiFi-and/ta-p/246605 https://community.cloudera.com/t5/Community-Articles/NIFI-1-x-For-Automatic-Music-Playing-Pipelines/ta-p/247994 https://community.cloudera.com/t5/Community-Articles/Using-Apache-NiFi-with-Apache-MXNet-GluonCV-for-YOLO-3-Deep/ta-p/248979 https://community.cloudera.com/t5/Community-Articles/Tracking-Air-Quality-with-HDP-and-HDF-Part-1-Apache-NiFi/ta-p/248265 https://community.cloudera.com/t5/Community-Articles/Monitoring-Energy-Usage-Utilizing-Apache-NiFi-Python-Apache/ta-p/247525 https://community.cloudera.com/t5/Community-Articles/Using-Command-Line-Security-Tools-from-Apache-NiFi/ta-p/248158 https://community.cloudera.com/t5/Community-Articles/Apache-NiFi-Processor-for-Apache-MXNet-SSD-Single-Shot/ta-p/249240 https://community.cloudera.com/t5/Community-Articles/Ingesting-Apache-MXNet-Gluon-Deep-Learning-Results-Via-MQTT/ta-p/248544 https://community.cloudera.com/t5/Community-Articles/Updating-The-Apache-OpenNLP-Community-Apache-NiFi-Processor/ta-p/248398 https://community.cloudera.com/t5/Community-Articles/Integration-Apache-OpenNLP-1-8-4-into-Apache-NiFi-1-5-For/ta-p/248010 https://community.cloudera.com/t5/Community-Articles/Tracking-Phone-Location-for-Android-and-IoT-with-OwnTracks/ta-p/244875 https://community.cloudera.com/t5/Community-Articles/Ingesting-Drone-Data-From-Ryze-Tello-Part-1-Setup-and/ta-p/249422 https://community.cloudera.com/t5/Community-Articles/Ingesting-RDBMS-Data-As-New-Tables-Arrive-Automagically-into/ta-p/246214 https://community.cloudera.com/t5/Community-Articles/Incrementally-Streaming-RDBMS-Data-to-Your-Hadoop-DataLake/ta-p/247927 https://community.cloudera.com/t5/Community-Articles/Ingesting-and-Analyzing-Street-Camera-Data-from-Major-US/ta-p/249194 https://community.cloudera.com/t5/Community-Articles/Basic-Image-Processing-and-Linux-Utilities-As-Part-of-a-Big/ta-p/249121 https://community.cloudera.com/t5/Community-Articles/Hosting-and-Ingesting-Data-From-Web-Pages-Desktop-and-Mobile/ta-p/244575 https://community.cloudera.com/t5/Community-Articles/QADCDC-Our-how-to-ingest-some-database-tables-to-Hadoop-Very/ta-p/245229 https://community.cloudera.com/t5/Community-Articles/Tracking-Air-Quality-with-HDP-and-HDF-Part-2-Indoor-Air/ta-p/249471 https://community.cloudera.com/t5/Community-Articles/Streaming-Ingest-of-Google-Sheets-with-HDF-2-0/ta-p/247764 https://community.cloudera.com/t5/Community-Articles/Ingesting-Golden-Gate-Records-From-Apache-Kafka-and/ta-p/247557 https://community.cloudera.com/t5/Community-Articles/Data-Processing-Pipeline-Parsing-PDFs-and-Identifying-Names/ta-p/249105 https://community.cloudera.com/t5/Community-Articles/Using-A-TensorFlow-quot-Person-Blocker-quot-With-Apache-NiFi/ta-p/248141 https://community.cloudera.com/t5/Community-Articles/Su-Su-Sussudio-Sudoers-Log-Parsing-with-Apache-NiFi/ta-p/249461 https://community.cloudera.com/t5/Community-Articles/Integrating-IBM-Watson-Machine-Learning-APIs-with-Apache/ta-p/247545 https://community.cloudera.com/t5/Community-Articles/Simple-Change-Data-Capture-CDC-with-SQL-Selects-via-Apache/ta-p/308376 https://community.cloudera.com/t5/Community-Articles/Deep-Learning-IoT-Workflows-with-Raspberry-Pi-MQTT-MXNet/ta-p/249456 https://community.cloudera.com/t5/Community-Articles/Parsing-Web-Pages-for-Images-with-Apache-NiFi/ta-p/248415 https://community.cloudera.com/t5/Community-Articles/Trigger-SonicPi-Music-Via-Apache-NiFi/ta-p/248587 https://community.cloudera.com/t5/Community-Articles/Using-Parsey-McParseFace-Google-TensorFlow-Syntaxnet-From/ta-p/246337 https://community.cloudera.com/t5/Community-Articles/Ingesting-osquery-Into-Apache-Phoenix-using-Apache-NiFi/ta-p/249308 https://community.cloudera.com/t5/Community-Articles/Converting-PowerPoint-Presentations-into-French-from-English/ta-p/248974 https://community.cloudera.com/t5/Community-Articles/Posting-Images-with-Apache-NiFi-1-7-and-a-Custom-Processor/ta-p/249017 https://community.cloudera.com/t5/Community-Articles/Parsing-Any-Document-with-Apache-NiFi-1-5-with-Apache-Tika/ta-p/247672

Really interested in partial updates, but haven't found any information on how physically the merges/upserts happen. It would be great if a doc like https://github.com/apache/kudu/blob/master/docs/design-docs/tablet.md existed for apache doris.