Hello. I have a question regarding SPICE.
My dataset source is coming from a RDS VIEW table. I have used this table in a dashboard, so that each time a user goes in, he/she will get the latest data from this VIEW table.
However, the dataset is getting larger and it takes longer to load the dashboard these days.
I know that SPICE speeds up the loading of data, but it seems to me that SPICE can only use it for data that are not constantly updated (that is VIEW table wouldn’t work).
Am I right about SPICE?
Hi @dixerio
The general idea in Business Intelligence is to enable users to interact with dashboard visuals to understand the business data, trends better. Users should be able to look at data in different ways, by filtering, slicing/dicing and/or drilling into details.
SPICE allows you to save data in QuicSight so the above mentioned interactions can be answered in the nick of time. Since this a store in QuickSight you need to refresh the SPICE dataset from time to time to bring update/new data from your underlying source.
Your approach of direct query may work for smaller datasets; but as you have seen now when data gets bigger querying the underlying data directly has implications. Each change in filter on the dashboard will need to be passed back to the database for an answer which is not ideal for users seeking quick answers.
SPICE datasets can be refreshed either fully or incremental (bring only new/updated data).
We use SPICE datasets for data that is changing every minute. We refresh our SPICE dataset every 15 minutes. We provide a footnote to the user indicating the freshness of the data by showing the last refreshed time of the dataset so the users are aware.
Regards,
Giri