21 karma · joined August 21, 2023
DeepComputing's blog on the matter, with an actual picture!
Edit: Higher resolution: https://deepcomputing.io/wp-content/uploads/2024/06/DC-ROMA_...
I'm linking the threads on the 11th gen and 12th gen Intel BIOS updates. It takes years for an update to be made available, and then it has endless problems with the update itself or after.
Nirav Patel writes from time to time they're gradually putting more effort into the maintenance of their older systems, but it doesn't instill confidence the older hardware will remain as well supported as alternatives.
11th gen: https://community.frame.work/t/11th-gen-intel-core-bios-3-10... 12th gen: https://community.frame.work/t/12th-gen-intel-core-bios-3-08...
Most business applications contain only the actual state of data, so by loading the data warehouse these changes can be saved and analyzed.
The simplest implementation is that each record gets a start_datetime and end_datetime so you can see the changes over time or figure out when something was deleted. Other techniques like data vault enable businesses to get a complete historic view of any given point in time.
If there is no soft deletion or audit table available, the alternative is to either 1) Select all the source primary keys to find out which no longer exist in the source but do exist in the data warehouse 2) If there is no primary key available, select the whole dateset for full row comparison
Since in corporate settings data warehouses load at least once a day, you can imagine the amounts of data being moved around needlessly.
For all my clients Snowflake is overkill when you look from the perspective of growth and scale. They'll never use that part of Snowflake. They might just do as well with DuckDB, Azure Synapse or any other analytical-oriented platform laying around.
What I do like with less-than-big use-cases is that (at least at Snowflake) you pay relatively little if you do relatively little data processing. It's not free, but it doesn't break the bank either.