getClusterPerf
Syntax
getClusterPerf([includeMaster=false])
Arguments
includeMaster (optional) indicates whether to include the controller information in the output.
Details
Returns configurations and performance measures for each node in the cluster. It
returns a table with the following columns:
| Field | Description | Unit |
|---|---|---|
| computeGroup | Compute group name | |
| host | Host name | |
| port | Port number | |
| site | (LAN) site | |
| mode | Node type or deployment mode. 0: data node;1: agent; 2: controller; 3: standalone mode; 4: compute node. | |
| state | A Boolean value indicating whether the node is alive. | |
| agentSite | Agent information of the current node. | |
| maxConnections | The maximum number of connections (from GUI, API, other nodes, etc.) to the local node. | |
| maxMemSize | The maximum memory allocated to DolphinDB. | GB |
| workerNum | The size of worker pool for regular interactive jobs. The default value is the number of CPU cores | |
| executorNum | The number of local executors. The default value is the number of CPU cores - 1 | |
| connectionNum | The number of connections to a local node. Note that the return value for the agent is a random number. | |
| name | Node alias | |
| memoryUsed | Memory used by the node | Bytes |
| memoryAlloc | Total memory allocated to DolphinDB on the node. | Bytes |
| cpuUsage | CPU usage | |
| avgLoad | Average CPU load | |
| maxRunningQueryTime | The maximum elapsed time of the queries that are currently running. | ns |
| runningJobs | The number of running jobs | |
| queuedJobs | The number of jobs in queue | |
| runningTasks | The number of running tasks | |
| queuedTasks | The number of tasks in queue | |
| jobLoad | CPU load of a job | |
| diskCapacity | Disk capacity | Bytes |
| diskFreeSpace | Available disk space | Bytes |
| diskFreeSpaceRatio | Available space ratio | |
| diskWriteRate | The rate at which data are written to disk. | Bytes/sec |
| diskReadRate | The rate at which data are read from disk. | Bytes/sec |
| lastMinuteWriteVolume | The data written to disk in the last minute. | Bytes |
| lastMinuteReadVolume | The data read from disk in the last minute. | Bytes |
| networkSendRate | The rate at which data are sent. | Bytes/sec |
| networkRecvRate | The rate at which data are received. | Bytes/sec |
| lastMinuteNetworkSend | Data sent in the last minute | Byte |
| lastMinuteNetworkRecv | Data received in the last minute | Byte |
| publicName | Returns publicName for a controller configured with publicName;Otherwise returns the MAC (Media Access Control) Address of the server for a controller, data node or agent. | |
| lastMsgLatency | Latency of the last received message | ns |
| cumMsgLatency | Cumulative latency of the messages | ns |
| isLeader | Whether it is the leader of a raft group. It only returns for a raft group of controllers. Note that it will return NULL for a raft group with crashed node. | |
| zone | The zone to which the node belongs |
The following measures are displayed only when the configuration parameter
perfMonitoring is set to 1.
| Field | Description | Unit |
|---|---|---|
| medLast10QueryTime | The median execution time of the previous 10 finished queries. | ns |
| maxLast10QueryTime | The maximum execution time of the previous 10 finished queries. | ns |
| medLast100QueryTime | The median execution time of the previous 100 finished queries. | ns |
| maxLast100QueryTime | The maximum execution time of the previous 100 finished queries. | ns |
Examples
getClusterPerf()