4 ms·
> it was basically untouchable because it was so slow. I've worked with a 5B row table on Snowflake with (maybe) no indexes, and while somewhat slow you could
by Simon_O_Rourke 2y ago
> it was basically untouchable because it was so slow.
I've worked with a 5B row table on Snowflake with (maybe) no indexes, and while somewhat slow you could still run reasonable queries on it in a minute or two.
- bhouston 2y agoI guess I liked the convenience with Bigquery of being able to do significant queries in a second or two on large data sets. It means I can do the query live for the user as they navigate the analytics dashboard and this simplifies design and implementation complexity significantly.