eG Monitoring
 

Measures reported by NetCluNodePerfTest

A node is a controller in a cluster. You can group pairs of nodes together to form a scalable cluster. Creating a cluster enables the nodes to pool their resources and distribute work across the cluster, while presenting administrators with a single entity to manage. Clustering also enables continuous service to end users if individual nodes go offline.

A cluster can contain up to 24 nodes (unless the iSCSI or FC protocols are enabled, in which case the cluster can contain up to eight nodes). Each node in the cluster can view and manage the same volumes as any other node in the cluster. The total file-system namespace, which comprises all of the volumes and their resultant paths, spans the cluster.

When new nodes are added to a cluster, there is no need to update clients to point to the new nodes. The existence of the new nodes is transparent to the clients.

Periodically monitoring the state and I/O activity of each of the node in the NetApp Cluster enables you to rapidly detect I/O overloads and figure out the nodes that are inconsistent/offline. This is exactly where the Node Performance test helps!

    This test auto-discovers the nodes on the NetApp Cluster, and periodically reports the following:

  • What is the current state of the node?
  • Which are the nodes that are busy processing I/O requests
  • Is the I/O load activity uniform across all the nodes? Are any nodes overloaded with I/O read-write requests?

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Status Indicates the current state of this node.   The values that this measure can report and their corresponding numeric values have been listed in the table below.
Measure Value Numeric Value
Unhealthy 0
Healthy 1

Note:

By default, this measure reports the above-mentioned Measure Values while indicating the current state of this node. However, in the graph of this measure, states will be represented using the corresponding numeric equivalents i.e., 0 or 1.

Total_ops Indicates the rate at which all operations were performed on this node. Ops/Sec  
Write_ops Indicates the rate at which the write operations were performed on this node. Ops/Sec A consistent decrease in the value of this measure could indicate a bottleneck when processing write requests. Compare the value of this measure across nodes to know which nodes are servicing write requests slowly.
Read_ops Indicates the rate at which the read operations were performed on this node. Ops/Sec A consistent decrease in the value of this measure could indicate a bottleneck when processing read requests. Compare the value of this measure across nodes to know which nodes service read requests slowly.
Http_ops Indicates the rate at which HTTP operations were performed on this node. Ops/Sec A consistent decrease in the value of this measure could indicate a bottleneck when processing HTTP requests. Compare the value of this measure across nodes to know which nodes service HTTP requests slowly.
Disk_data_read Indicates the rate at which data was read from the disk of this node. MB/Sec Comparing the value of this measure across nodes will help you identify the node that is the slowest in terms of reading the data from the disk.
Net_data_received Indicates the rate at which network data is received on this node. MB/Sec  
Net_data_sent Indicates the rate at which network data is sent through this node. MB/Sec  
Avg_system_latency Indicates the average time taken by the system to perform operations through this node. Secs A high value for this measure is a cause of concern.