ApsaraDB for ClickHouse provides various metrics for you to monitor the status of ApsaraDB for ClickHouse clusters in real time. This helps you identify issues at the earliest opportunity and optimize the cluster performance. This topic describes the metrics of ApsaraDB for ClickHouse clusters.
Metrics of ApsaraDB for ClickHouse clusters
The following table describes the metrics of clusters purchased after December 1, 2021.
For more information about the metrics of clusters purchased before December 1, 2021, see the Metrics of old-version clusters section of this topic.
Community-compatible Edition
Metric (console) | Unit | Description |
Disk throughput | Bytes | The read and write throughput (bandwidth) of the disk. |
Disk IOPS | Requests/s | The number of read and write operations performed per second on the disk. |
Disk usage | Bytes | The amount of used disk space. |
Disk Usage-Single Node Statistics | Bytes | The total amount of used disk space for a single node. |
Disk usage | Percentage (%) | The ratio of the used disk space to the maximum amount of available disk space. |
Disk Usage-Single Node Statistics | Percentage (%) | The ratio of the used disk space to the maximum amount of available disk space for a single node. |
CPU usage | Percentage (%) | The average CPU utilization of each node in the cluster. |
inode usage | Percentage (%) | The ratio of the number of used inodes to the maximum number of available inodes. Note Inodes are used to identify files and directories in a Linux operating system. |
Memory usage | Bytes | The amount of memory that is used by each node of the cluster. |
Memory usage | Percentage (%) | The average memory usage of each node in the cluster. |
Data Part number | Counts | The total number of data shards. |
Number of Inactive Data Parts | Counts | The number of inactive data shards. |
ZooKeeper CPU Usage | Percentage (%) | The average CPU utilization of the nodes in the ZooKeeper cluster. |
ZooKeeper memory usage | Percentage (%) | The average memory usage of the nodes in the ZooKeeper cluster. |
Write size per second | Bytes | The amount of data written to each node per second in the cluster. |
Number of rows written per second | Rows/s | The number of rows written to each node per second in the cluster. |
QPS | Counts | The number of queries processed per second. |
TPS | Counts | The number of transactions processed per second. |
Network throughput | Bytes | The network bandwidth. |
Number of TCP connections | Counts | The number of TCP connections for the cluster. |
Number of HTTP connections | Counts | The number of HTTP connections for the cluster. |
Number of Init Query runs | Counts | The number of running statements for non-forwarded queries. |
Number of Query runs | Counts | The number of running query statements. |
Number of failed queries | Counts | The number of failed query statements. |
Number of failed Insert queries | Counts | The number of failed INSERT statements. |
Number of delayed Insert | Counts | The number of INSERT statements that have latencies. |
Number of failed Select queries | Counts | The number of failed query statements. |
Number of Merge runs | Counts | The number of running merge tasks. |
Number of running Mutation | Counts | The number of running data modification tasks. |
Number of temporary files in distributed tables | Counts | The number of temporary files in the distributed tables. |
Number of MaterializeMySQL | Counts | The number of databases created by using MaterializeMySQL to synchronize data from MySQL to ApsaraDB for ClickHouse. |
Cold storage usage | Bytes | The storage space used for storing cold data. |
Number of MaterializeMySQL synchronization failures | Counts | The number of synchronization failures of the databases created by using MaterializeMySQL. |
The number of errors in Kafka's appearance consumption. | Counts | The number of consumed Kafka external tables that failed to be synchronized. |
Zookeeper leader | - | The leader of the ZooKeeper cluster. The node whose metric value is set to 1 is the leader. |
Kafka appearance number | Counts | The number of Kafka external tables that are created. |
Amount of node queue accumulation | Counts | The number of data packets sent by ApsaraDB for ClickHouse nodes to ZooKeeper in the send queue. |
Zookeeper average delay | Milliseconds | The average latency of ZooKeeper nodes. |
Number of Zookeeper Node | Counts | The number of ZooKeeper znodes. |
ZookeeperTPS | Requests/s | The transactions per second (TPS) of ZooKeeper. |
Zxid Usage | Percentage (%) | The ZooKeeper transaction ID (zxid) usage. When the zxid usage reaches 100%, a new leader is selected. |
Enterprise Edition
Metric (console) | Unit | Description |
CPU usage | Percentage (%) | The average CPU utilization of each node in the cluster. |
CCU | Counts | The amount of used computing resources for the cluster. |
Memory usage | Bytes | The amount of memory that is used by each node of the cluster. |
Memory usage | Percentage (%) | The average memory usage of each node in the cluster. |
OSS Usage | Bytes | The amount of used storage resources for the cluster. |
Data Part number | Counts | The total number of data shards. |
Number of Inactive Data Parts | Counts | The number of inactive data shards. |
Keeper CPU Usage | Percentage (%) | The CPU utilization of Keeper for ApsaraDB for ClickHouse. |
Keeper Memory Usage | Bytes | The amount of memory that is used by Keeper of ApsaraDB for ClickHouse. |
Keeper Memory Usage | Percentage (%) | The memory usage of Keeper for ApsaraDB for ClickHouse. |
Write size per second | Bytes | The amount of data written to each node per second in the cluster. |
Number of rows written per second | Rows/s | The number of rows written to each node per second in the cluster. |
Network throughput | Bytes | The read and write throughput (bandwidth) of the disk. |
OSS network throughput | Bytes | The read and write throughput (bandwidth) of Object Storage Service (OSS). |
OSS IOPS | Requests/s | The number of read and write operations performed per second by OSS. |
TPS | Requests/s | The number of transactions processed per second. |
QPS | Requests/s | The number of queries processed per second. |
Number of Init Query runs | Counts | The number of running statements for non-forwarded queries. |
Number of Query runs | Counts | The number of running query statements. |
Number of failed queries | Counts | The number of failed queries on the node. |
Number of failed Insert queries | Counts | The number of failed INSERT statements. |
Number of delayed Insert | Counts | The number of INSERT statements that have latencies. |
Number of failed Select queries | Counts | The number of failed SELECT queries on the node. |
Number of running Mutation | Counts | The number of running data modification tasks. |
Number of Merge runs | Counts | The number of running merge tasks. |
Number of TCP connections | Counts | The number of TCP connections for the cluster. |
Number of HTTP connections | Counts | The number of HTTP connections for the cluster. |
Metrics of old-version clusters
Old-version clusters are ApsaraDB for ClickHouse Community-compatible Edition clusters purchased before December 1, 2021.
Metric (console) | Unit | Description |
CPU Usage | Percentage (%) | The CPU utilization of each node in the cluster. |
Memory Usage | Percentage (%) | The memory usage of each node in the cluster. |
Memory Usage | Bytes | The amount of memory that is used by each node of the cluster. |
Disk Usage Ratio | Percentage (%) | The ratio of the used disk space to the maximum amount of available disk space. |
Disk Usage | Bytes | The amount of used disk space. |
Disk IOPS | Requests/s | The number of read and write operations performed per second on the disk. |
Disk IOPS | Bytes | The size of data read from and written to the disk per second. |
Database Connection Usage | Percentage (%) | The ratio of the number of established database connections to the maximum number of available database connections. |
Database Connections | Counts | The number of established database connections. |
TPS | Requests/s | The number of transactions processed per second. |
Rows Written Per Second | Rows/s | The number of rows written to each node per second. |
Size of Data Written per second | Bytes | The size of data written to each node per second. |
QPS | Requests/s | The number of queries processed per second. |
Average ZooKeeper wait time | Milliseconds | Indicates the responsiveness of ZooKeeper. |
Average I/O wait time | Milliseconds | Indicates the I/O responsiveness. |
Average CPU wait time | Milliseconds | Indicates the responsiveness of the CPU. |