Redis CPU utilization spikes during cache key deletions

Cache key deletion operations cause Redis CPU utilization to spike to 83%+ because Redis must scan the entire key table to find matching keys to delete. This creates a table-scan scenario that severely degrades performance under load, with CPU utilization climbing steadily as traffic increases.

Remove or minimize cache key deletion operations from application code. Implement alternative invalidation strategies such as time-based expiration or timestamp-based invalidation metadata. Monitor EngineCPUUtilization and CPU Utilization metrics together. If deletions are necessary, use pipelining with HiredisParser for batch operations. Test in dedicated performance environment before deployment.