We want to create a predicate index that supplements the ACORN predicate-agnostic index. For the MVP, we will hardcode this logic by explicitly initializing a special index. Provide the user an API to create a predicate index before the experiment based on the findings from #30.
In order to create a generalizable approach, it might make sense to create our predicate index during the experiment before issuing any workload. This would require parameterizing resources (CPU and memory), but this is deeply subject to change.
We want to create a predicate index that supplements the ACORN predicate-agnostic index. For the MVP, we will hardcode this logic by explicitly initializing a special index. Provide the user an API to create a predicate index before the experiment based on the findings from #30.
In order to create a generalizable approach, it might make sense to create our predicate index during the experiment before issuing any workload. This would require parameterizing resources (CPU and memory), but this is deeply subject to change.