Background
I am planning to use GCP's Bigtable (BT) offering for one of my projects. This is a NOT_SO_CRITICAL use-case, hence i am exploring a single cluster setup (COST factor).
Question
What are the intra-zone High Availability (HA) gurantees for Bigtable ?
I am ok for any outages which occur at the zone level (relatively rare event, acceptable for my use-case). Concern for me is BT causing outages incase of single machine going down as part of ad-hoc maintenance.
PS : Google only allows a single big-table cluster (per BT instance) within an availability zone. This makes me believe even more that there would be some HA guarantees within a zone, but not getting any documentation around this.
What are the intra-zone High Availability (HA) gurantees for Bigtable ?
The Covered Service will provide a Monthly Uptime Percentage to Customer as follows (the "Service Level Objective" or "SLO"):
Cloud Bigtable - Zonal instance (single cluster) >= 99.9%