QPS
| Metric | Description | Unit |
|---|---|---|
| sql_all_count | The number of SQL statements processed per second. | times/s |
| sql_delete_count | The number of DELETE statements processed per second. | times/s |
| sql_insert_count | The number of INSERT statements processed per second. | times/s |
| sql_other_count | The number of other statements processed per second, such as DDL, DCL, and DTL statements. | times/s |
| sql_replace_count | The number of REPLACE statements processed per second. | times/s |
| sql_select_count | The number of SELECT statements processed per second. | times/s |
| sql_update_count | The number of UPDATE statements processed per second. | times/s |
Response time
| Metric | Description | Unit |
|---|---|---|
| sql_all_rt | The average response time for SQL statements. | ms |
| sql_delete_rt | The average response time for DELETE statements. | ms |
| sql_insert_rt | The average response time for INSERT statements. | ms |
| sql_other_rt | The average response time for other statements, such as DDL, DCL, and DTL statements. | ms |
| sql_replace_rt | The average response time for REPLACE statements. | ms |
| sql_select_rt | The average response time for SELECT statements. | ms |
| sql_update_rt | The average response time for UPDATE statements. | ms |
| sql_all_rt_p99 | The average response time for SQL statements at the 99th percentile. | ms |
| sql_delete_rt_p99 | The 99th percentile response time for DELETE statements. This value indicates that 99% of the DELETE statements have a response time less than or equal to this value. This metric is used to measure the performance of slow queries in the database. | ms |
| sql_insert_rt_p99 | The 99th percentile response time for INSERT statements. This value indicates that 99% of the INSERT requests have a response time less than or equal to this value. The unit is ms. | ms |
| sql_other_rt_p99 | The 99th percentile response time for other types of SQL statements. The unit is ms. | ms |
| sql_replace_rt_p99 | The 99th percentile response time for REPLACE statements. The unit is ms. | ms |
| sql_select_rt_p99 | The 99th percentile response time for SELECT statements. This value indicates that 99% of the SELECT queries have a response time less than or equal to this value. The unit is ms. | ms |
| sql_update_rt_p99 | The 99th percentile response time for UPDATE statements. This metric is used to measure the tail latency of UPDATE statements in the database. The unit is ms. | ms |
| sql_all_rt_p95 | The average response time for SQL statements at the 95th percentile. | ms |
| sql_delete_rt_p95 | The 95th percentile response time for DELETE statements. This metric is used to measure the latency of DELETE statements in the database. | ms |
| sql_insert_rt_p95 | The 95th percentile response time for INSERT statements. The unit is ms. | ms |
| sql_other_rt_p95 | The 95th percentile response time for other types of SQL statements. The unit is ms. | ms |
| sql_replace_rt_p95 | The 95th percentile response time for REPLACE statements. This metric is used to measure the response time of REPLACE requests in the database. The unit is ms. | ms |
| sql_select_rt_p95 | The 95th percentile response time for SELECT statements. This value indicates that 95% of the SELECT queries have a response time less than or equal to this value. This metric is used to measure the tail latency of SELECT queries in the database. The unit is ms. | ms |
| sql_update_rt_p95 | The 95th percentile response time for UPDATE statements. The unit is ms. | ms |
| sql_all_rt_p90 | The average response time for SQL statements at the 90th percentile. | ms |
| sql_delete_rt_p90 | The 90th percentile response time for DELETE statements. This value indicates that 90% of the DELETE statements have a response time less than or equal to this value. This metric is used to measure the stability of DELETE operations in the database. | ms |
| sql_insert_rt_p90 | The 90th percentile response time for INSERT statements. This value indicates that 90% of the INSERT requests have a response time less than or equal to this value. This metric is used to measure the tail latency of INSERT operations in the database. | ms |
| sql_other_rt_p90 | The 90th percentile response time for other types of SQL statements. This value indicates that 90% of the queries have a response time less than or equal to this value. This metric is used to measure the tail latency of queries in the database. | ms |
| sql_replace_rt_p90 | The 90th percentile response time for REPLACE statements. This metric is used to measure the response time of slow queries. | ms |
| sql_select_rt_p90 | The 90th percentile response time for SELECT statements. This value indicates that 90% of the SELECT queries have a response time less than or equal to this value. This metric is used to measure the tail latency of SELECT queries in the database. | ms |
| sql_update_rt_p90 | The 90th percentile response time for UPDATE statements. The unit is ms. | ms |
Average IOPS
| Metric | Description | Unit |
|---|---|---|
| io_count | The number of I/O operations performed by the SSStore per second. | times/s |
| io_read_count | The number of read operations performed by the SSStore per second. | times/s |
| io_write_count | The number of write operations performed by the SSStore per second. | times/s |
I/O latency
| Metric | Description | Unit |
|---|---|---|
| io_rt | The average I/O latency. | μs |
| io_read_rt | The average read latency of the SSStore. | μs |
| io_write_rt | The average write latency of the SSStore. | μs |
I/O throughput
| Metric | Description | Unit |
|---|---|---|
| io_size | The amount of data processed by the SSStore per second. | byte |
| io_read_size | The amount of data read by the SSStore per second. | byte |
| io_write_size | The amount of data written by the SSStore per second. | byte |
Non-idle wait time of a tenant database
| Metric | Description | Unit |
|---|---|---|
| ob_waitevent_count | The number of wait events in a tenant database. This metric is used to monitor the frequency of wait events in the database to locate performance bottlenecks. | times |
| ob_foreground_waitevent_count | The number of wait events in the foreground of a tenant database. This metric is used to monitor the frequency of wait events in the foreground sessions of OceanBase Database. | times/s |
| ob_background_waitevent_count | The number of wait events in the background of a tenant database. This metric is used to monitor the wait events of background tasks in OceanBase Database. | times |
Number of execution events in each state
| Metric | Description | Unit |
|---|---|---|
| ob_sql_event | The number of SQL execution events in each state. This metric monitors the distribution of SQL execution states in OceanBase Database. | times |
| ob_sql_event_in_parse | The number of SQL execution events in the parse state. This metric monitors the concurrent load in the SQL parsing stage. | count |
| ob_sql_event_in_pl_parse | The number of SQL execution events in the PL parsing state. This metric monitors the concurrent execution in the PL/SQL parsing stage. | count |
| ob_sql_event_in_plan_cache | The number of SQL execution events in the plan cache state. This metric monitors the number of SQL execution plans cached. | count |
| ob_sql_event_in_sql_optimize | The number of SQL execution events in the SQL optimization state. This metric monitors the concurrent or queued SQL optimization. | count |
| ob_sql_event_in_sql_execution | The number of SQL execution events in the SQL execution state. This metric monitors the number of SQL events being executed. | count |
| ob_sql_event_in_px_execution | The number of SQL execution events in the parallel execution (PX execution) state. This metric monitors the concurrency of parallel queries. | count |
| ob_sql_event_in_sequence_load | The number of SQL execution events in the sequence load state. This metric monitors the concurrency of sequence loading operations. | count |
Number of sessions
| Metric | Description | Unit |
|---|---|---|
| all_session | The number of current sessions. This metric indicates the number of active connections to the database. | count |
| active_session | The number of active sessions. | count |
TPS
| Metric | Description | Unit |
|---|---|---|
| transaction_count | The number of transactions processed per second. | times/s |
| transaction_commit_count | The number of committed transactions per second. | times/s |
| transaction_rollback_count | The number of rolled back transactions per second. | times/s |
| transaction_timeout_count | The number of timed-out transactions per second. | times/s |
Average transaction processing time on the server
| Metric | Description | Unit |
|---|---|---|
| transaction_rt | The average transaction processing time on the server. | ms |
| transaction_commit_rt | The average transaction commit time. | ms |
| transaction_rollback_rt | The average transaction rollback time. | ms |
Request wait queue time
| Metric | Description | Unit |
|---|---|---|
| request_queue_time | The wait time of an SQL request in the wait queue. | μs |
Number of committed transaction logs per second
| Metric | Description | Unit |
|---|---|---|
| trans_commit_log_count | The number of committed transaction logs per second. | times/s |
Size of committed transaction logs per second
| Metric | Description | Unit |
|---|---|---|
| clog_trans_log_total_size | The size of committed transaction logs per second. | byte |
Number of transactions
| Metric | Description | Unit |
|---|---|---|
| transaction_partition_count | The number of transactions per second. | times/s |
| transaction_multi_partition_count | The number of distributed transactions per second. | times/s |
| transaction_single_partition_count | The number of ordinary transactions per second. | times/s |
Average network synchronization time for each transaction log
| Metric | Description | Unit |
|---|---|---|
| trans_commit_log_sync_rt | The average network synchronization time for each transaction log. | ms |
Database work time of the tenant
| Metric | Description | Unit |
|---|---|---|
| ob_worktime | The work time of the tenant database. | s |
| ob_foreground_worktime | The foreground work time of the tenant database. | s |
| ob_background_worktime | The background work time of the tenant database. | s |
Non-idle wait time of the tenant database
| Metric | Description | Unit |
|---|---|---|
| ob_no_idle_waiting_time | The non-idle wait time of the tenant database. | s |
| ob_foreground_no_idle_waiting_time | The foreground non-idle wait time of the tenant database. | s |
| ob_background_no_idle_waiting_time | The background non-idle wait time of the tenant database. | s |
Total size of the tenant log disk
| Metric | Description | Unit |
|---|---|---|
| ob_tenant_log_disk_total_bytes | The total size of the tenant log disk. | GiB |
Used size of the tenant log disk
| Metric | Description | Unit |
|---|---|---|
| ob_tenant_log_disk_used_bytes | The used size of the tenant log disk. | GiB |
Data occupied by the tenant
| Metric | Description | Unit |
|---|---|---|
| ob_tenant_server_required_size | The data occupied by the tenant. | GiB |
Data size of the tenant
| Metric | Description | Unit |
|---|---|---|
| ob_tenant_server_data_size | The data size of the tenant. | GiB |
Size of binlog disk used
| Metric | Description | Unit |
|---|---|---|
| ob_tenant_binlog_disk_used | The size of the binlog disk used. | GiB |