Databricks: Melting the SNOW

This is a snippet of the transcript, sign up to read more.

How would you describe Databricks; how it’s doing and how is the culture?

I saw that in your questions. I would say the culture is pretty amazing. I've enjoyed it; I've always compared company culture to what I experienced at Tableau. Tableau was a founder-led organization with Christian Chabot, Chris Stolte, and some of those folks who were just very welcoming. It was you're on a mission to do missionary selling on self-service BI, taking out legacy Cognos, legacy BusinessObjects, those types of vendors.

This is a snippet of the transcript, sign up to read more.

Could you take a step back and explain to a layperson what Databricks does? How it is in relationship with the hyperscalers and the public clouds, and how does it compete with Snowflake? My initial understanding was that the use cases were different and complementary, but that seems to be changing.

Essentially, our go-to-market is talking about the lakehouse. The lakehouse combines the data lake, which has all your very granular information to power your machine learning models; you need the lowest level of data, the click, the byte. Then a combination of data warehouse, which is aggregate data for the business users and business analysts to analyze in Tableau in their reporting solutions because they're not going to be able to make sense of a huge wealth of data, they need curated data. In the past, you had Hadoop for the data lake, as a cheap file store to store this massive amount of data; it's hard to retrieve and had to have a separate governance model. Then you had the data warehouse, which was your curated data. Again, that was more expensive, so you minimized how much data was in there, and you had to have another governance model in there, too, that's your Redshift, Snowflake, BigQuery, Teradata, and the list goes on.

This is a snippet of the transcript, sign up to read more.

Is the lakehouse where you started?

We started at more of a data science workbench because our founders created Apache Spark. Apache Spark is the de facto way to process big data. We started by just being the data scientists' tool for processing large amounts of data and doing that faster and more efficiently than Amazon EMR, which is just a version of Spark. That's where we started. Over time, with this core technology of processing big amounts of data, we've been able to, what I would say, shift right more. Shift right means doing more user-friendly ETL, extracting transform load workloads, and doing more data warehousing workloads. So that's where we're headed and where we are.

This is a snippet of the transcript, sign up to read more.