Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
blueglassfish
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
1.
▲
by
blueglassfish
5y ago
That's a respectable amount for a DW, true. Spark and it's ilk are designed for much larger scales though. Multiple FAANG use cases for Spark are in the petabytes per week range.
2.
▲
by
blueglassfish
5y ago
I’m not sure this locks anyone in. The APIs are open and Spark code will run on, say EMR, just fine. Every vendor, be it Snowflake, Databricks, EMR, Athena, BQ, … charges for use of the engine. The difference with a Lakehouse is that one do