Optimization recommendations for Mimir Setup - ~300M Active Time Series

[AstroEngineeer]

Ingestion: ~300M Active Time Series
Tenant: Single
Replication Factor: 1
Zone Replication: False
Deployment Env: Kubernetes Cluster - GKE

My Helm Vaules.yaml

I've used the planning capacity doc to estimate the resource usage and few other references like

1. Scaling ingester above 1,500,000 in-memory series each? #3287
2. Scaling / Configuration recommendations for compactor #3369

i understand that the rollout might take a very long time since the ingester are big. Any advice to reduce this?

I would greatly appreciate any optimization recommendations for this setup.