Redis SCAN performance concern from Pods cache group flush (e.g. pods_post_type_

Hello Pods Team,

We’ve recently observed performance issues caused by extensive SCAN commands hitting our Redis instance. Upon investigation, it appears that the Pods plugin is performing wildcard key scans on patterns like:

MATCH wp_*:pods_post_type_storage__pods_group:**

These operations are initiated frequently and result in significant Redis CPU load, especially under high-traffic conditions.

We suspect this may be part of the cache invalidation logic (possibly within pods_cache_flush() or pods_cache_flush_group()), which scans Redis keys matching certain group patterns and deletes them. The full key scan seems to be implemented using Redis’s SCAN command, likely via a direct Redis client (Predis or PhpRedis).

We would like to ask:

1. Is this behavior intended or configurable?
   Can we limit or disable full key scans related to Pods group cache flushes?
2. Is there a recommended way to prevent pods_post_type_storage__pods_group keys from being scanned like this, or alternatively, can group-level cache invalidation be avoided?
3. Could you consider adding a more efficient mechanism (e.g., cache tagging, or selective key deletion) in future versions to avoid full keyspace scans in Redis?

This issue is currently impacting Redis performance on our production site (running at scale), and we’d appreciate any guidance or patches you can offer to mitigate this.

Thank you in advance for your time and support.

Best regards,

Developer Mook