Users report the on-premises Instana backend UI is extremely slow, with dashboards taking 45 seconds or more to render. All backend processes are running. The number of monitored entities grew from 500 to 5,000 over the past month. What should the administrator investigate first to address the performance degradation?

Instana on-premises backend resource requirements scale significantly with the number of monitored entities, and a 10x growth in entity count dramatically increases data processing, storage write throughput, and query demands. The official Instana sizing documentation specifies CPU, RAM, and disk I/O recommendations per entity count tier. If the backend servers were originally sized for 500 entities and have not been scaled, performance degradation is expected and the remedy is infrastructure scaling. The correct answer is 'Review CPU utilization, available RAM, and disk I/O on the Instana backend servers against the official sizing guidelines for 5,000 entities, and scale the backend infrastructure if resources are insufficient'. The incorrect options — 'Clear the browser cache and local storage on affected users workstations to remove stale UI data', 'Direct all Instana agents to increase their reporting intervals to reduce inbound data volume to the backend', 'Archive unused dashboards to lower the concurrent query load on the backend data store' — are wrong because they do not align with IBM Instana's architecture or recommended practices for this scenario. Understanding this concept is essential for the Domain 6: Troubleshooting domain of the IBM Instana Observability certification.