In some scenarios, performance bottlenecks may occur due to load imbalance between nodes. You can run the following command to query the distribution of SQL traffic and QPS among OBServer nodes:
obclient> SELECT/*+ PARALLEL(15)*/t2.zone, t1.svr_ip, COUNT(*) AS RPC_COUNT,
AVG(t1.elapsed_time), AVG(t1.queue_time)
FROM oceanbase.GV$OB_SQL_AUDIT t1, __all_server t2
WHERE t1.svr_ip = t2.svr_ip
AND tenant_id = 1001
AND SQL_ID = 'BF7AA13A28DF50BA5C33FF19F1DBD8A9'
AND IS_EXECUTOR_RPC = 0
AND request_time > (time_to_usec(now()) - 1000000)
AND request_time < time_to_usec(now())
GROUP BY t1.svr_ip;