TiDB doesn't want to solve all the problems, no silver bullet, right? And I think it's all about workload, some analytical workload requires CPU/Network resource instead of I/O, and OLAP workload isn't that frequently. Storage layer and computing layer are separated in TiDB stack, I think it's possible for some workload.
The problem is that OLAP is run by analysts/business/data.science people and they may mess up with their queries/workloads. without isolation the production performance could impact the user expirience
