[System Test] CPU constantly at ~95% on orchestrator node and some other KV nodes

Build : 7.0.0-4955
Test : -test tests/integration/cheshirecat/test_cheshirecat_kv_gsi_coll_xdcr_backup_sgw_fts_itemct_txns_eventing_cbas_scale3.yml -scope tests/integration/cheshirecat/scope_cheshirecat_with_backup.yml
Scale : 3
Iteration : 3rd

In the longevity test, the orchestrator node is 172.23.108.103. The CPU on this node is constantly very high - 90%+ all the time. See this chart for CPU util over the week -

Output of top on 172.23.108.103 :

Due to this, we see timeout exceptions in test scripts, and UI is practically unresponsive if accessed via 172.23.108.103:8091. If the UI is accessed from other nodes, it is much more responsive.

Other KV nodes too experience high CPU usage over a prolonged period of time.

172.23.106.100

Output of top on 172.23.106.100

172.23.121.3

172.23.97.121

I have attached the current latest logs, but logs are available for the past 5 days if required. Please let me approximate timestamp for the logs required, and I will provide the nearest set of logs.