Evaluate Performance in Detail › Tabular Report Sections › Cluster Disk and Server Statistics (with By Node Breakout) › Disk By Cluster Format

Disk By Cluster Format

In the By Cluster format for disk statistics, the leftmost column of the tabular report contains the name of each disk that is accessible cluster-wide. Disk names are usually prefixed by a node name or allocation class. Each disk name begins a line of disk statistics. Each statistic shows the activity for the specific disk due to all nodes in the cluster.

When determining cluster-wide statistics for a disk, only statistics from nodes with direct access are considered. This is because I/O for nodes with remote access is processed by nodes directly accessing the disk and is therefore already included in the node with direct access statistics. For example, the software would calculate cluster rates for disks DU1 and DB2 in the following figure as follows:

• For disk DU1 with the HSC, all I/Os come to this disk through host (direct access) nodes A and B. Therefore, adding the I/O statistics from each of these nodes provides the total cluster I/O for disk DU1. The I/O statistics of node C are already included in the I/O statistics for nodes A and B and are therefore ignored.
• For disk DB2, node C is the host (direct access) node. All I/Os come to this disk through node C; therefore, the I/O statistic of node C provides the total cluster I/O rate for disk DB2. As above, the I/O statistics of nodes A and B are already included in the I/O statistics for node C and are therefore ignored.

Example Configuration: Disk By Cluster Format