Scott Teal

💯💯💯

@chimerasaurus youtu.be/1-APptl8URs?t=…

@Cloudflare @CloudianStorage @Dell @HitachiGlobal @Minio @PureStorage @scality @wasabi_cloud Of course, you can run SQL queries on files like parquet, avro, orc, csv, and json stored in S3-compatible storage using External Tables.

@Cloudflare @CloudianStorage @Dell @HitachiGlobal @Minio @PureStorage @scality @wasabi_cloud Snowflake’s Document AI in private preview isn’t only just for documents in Snowflake. You can do inference on files in these S3-compatible storage devices too!

You can connect @SnowflakeDB to s3-compatible storage devices to process and query files! Demo by @pm_saurin youtu.be/7EZqXydceGU

@mim_djo @RonOrtloff @chimerasaurus @SnowflakeDB This demo used an “unmanaged” Iceberg table, so Athena = writer, and Snowflake = reader via Glue catalog integration. “Managed” Iceberg tables enable read+write from Snowflake, Spark can read like so snowflake.com/blog/iceberg-t… As the top chart shows, much less perf tradeoff

@RonOrtloff @scottteal @chimerasaurus @SnowflakeDB Let's be annoying a bit 😁, the only writer supported is ❄️ ?

And files are stored in the bucket I provide, which in this case is S3

#ApacheSpark reading #ApacheIceberg table created by @SnowflakeDB An early first step which should give you a sense where we're headed. More to come!

@chimerasaurus Dataproc too

I just published "Snowflake tables on Cloudflare R2". When replicating data to multiple regions, egress costs can be quite significant. So this is exciting: Now @SnowflakeDB can use @Cloudflare R2 as external storage for tables— featuring 0 egress fees❄️ @hoffa/snowflake-tables-on-cloudflare-r2-b5496c4ae9ec" target="_blank" rel="nofollow noopener">medium.com/@hoffa/snowfla…

1/Excellent presentation how #SnowflakeDB is implementing #ApacheIceberg, TL;DR, the data is saved as parquet in the user Cloud storage, but for efficiency reason, the metadata is generated on the fly when you connect your tool to the Snowflake Catalog youtube.com/watch?v=Kr-Vzv…

Interested in AI and data-led decisions? Check out our latest Success Story to learn how @hyperfinity + @SnowflakeDB are helping businesses use machine learning to make decisions related to areas like smarter supply chains, optimized pricing, and more: bit.ly/3Y61tGj

Neat story of a migration from Hive tables of JSON to Iceberg tables of Parquet at @Shopify Thanks for sharing, Marc! youtu.be/nJBBw-xnLU8 #ApacheIceberg

@SnowflakeDB 💙 Python + @streamlit

@rajkrsen @neelesh_salian Similarly, my first thought was a table format like Apache Iceberg if you seek more control and interoperability of files in S3. Of course, that gained control and interoperability comes with the added responsibilities of security and maintenance.

@neelesh_salian Isn’t that Delta Lake? Trying to understand why you think it’s not

Can we just have a Data warehouse with Parquet files in S3 and then build stuff around it? Build whatever you need for BI and other data access. #data #warehouse

@TableauTim @SnowflakeDB @awscloud I love it ❄️ There are far more expert people out there compared to me, like the Snowflake Data Superheroes community.snowflake.com/s/dataheroes ...but if you ever have q's or ideas, let me know! I recently completed the Data Lake Workshop, covers some similar topics learn.snowflake.com/tracks

Pushing forward with more @SnowflakeDB content as I learn. The new video looks at how to load data from @awscloud S3 into Snowflake. It's an end-to-end tutorial, including the bumps and troubleshooting when my SQL didn't work. youtu.be/DV8EbD9s_CM

But these are just my favorites! There are many more sessions available on-demand, and more streaming live tomorrow snowflake.com/build/agenda/

Last but not least, STREAMING This session provides a nice overview and use cases for various ingestion methods with Snowflake. There’s also a demo of the Kafka Connector with Snowpipe Streaming snowflake.com/build/agenda?a…

Sharing some personal favs from day 1 of #SnowflakeBUILD (in no particular order) First, a @felipehoffa sighting! Here's an interesting convo with a few engineering and product leaders across a few different parts of @SnowflakeDB snowflake.com/build/agenda?a…