MicrosoftAn additional option to update your gold layer tables
Update policies and materialized views can be quite handy if the gold layer tables are time-critical. The disadvantage is that they might consume a lot of performance. If this is done multiple times, it might slow down your Kusto cluster and finally increase your costs. That's why I would like to mention an alternate approach.
Sometimes reporting users may be happy to get regular updated data of, let's say four or six hours or even one day. Then, it might be a better option to avoid using update policies or materialized views and instead create a simple, scheduled Data Factory job that appends your new data to the gold layer: .append gold_layer_table <| fromSilverToGold(lastTimestamp, currentTimestamp). The disadvantage here is that you need a Data Factory instance in addition to your Kusto cluster.
As a summary, both ways are simple and both ways will do the job in a perfect manner. If you are working in a complex Data Lakehouse infrastructure anyway, your data architects may decide which option fits best - depending on the business case, requirements and budget.
