Metrics storage spike due to host tag (PR #7530)

[neil-xie]

Description

We’re seeing a massive increase in metrics storage usage due to the host tag introduced in #7530. Our team is actively developing a resolution, I will update this thread once the fix is merged.

Steps to Reproduce / How to Trigger

Introduced in v1.3.7-prerelease25

Expected Behavior

Massive increase in metrics storage usage

Actual Behavior

What actually happened.

Logs / Screenshots

Relevant logs, stack traces, workflow history, or screenshots.

Environment

• Cadence server version: v1.3.7-prerelease25
• Cadence SDK language and version (if applicable): Language and SDK version
• Cadence web version (if applicable): Web UI version
• DB & version: Database type and version
• Scale: Cluster size, workload volume, or other scale details

[neil-xie]

Reverted in #7675