eG Monitoring
 

Measures reported by NetCluDiskPerfTest

For an aggregate to perform effectively, there must be adequate free space in the aggregate. Periodical monitoring of the aggregates for the space usage of each of the aggregates is mandatory to identify probable space contentions which may indirectly slowdown the performance of the aggregates. This test auto-discovers the aggregates in the NetApp Cluster and reports the space utilization of each aggregate.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Cpreads Indicates the rate at which the read requests from the user are serviced during a Consistency Point (CP) operation in this disk. Reads/Sec A consistent decrease in the value of this measure could indicate that CP operations are slowing down the processing of read requests.
Cpreads_latency Indicates the time taken for retrieving data or metadata associated with user requests during a Consistency Point operation in this disk. Secs  
Disk_busy Indicates the percentage of time there was atleast one outstanding request (i.e., read or write) to this disk. Percent A value greater than 70% is a cause of concern which indicates performance degradation of the disk.

Comparing the percentage of time that the different disks are busy, an administrator can determine whether the application load is properly balanced across the different disks.
IO_pending Indicates the average number of I/O requests to this disk that were pending processing. Number A low value is desired for this measure. A gradual/sudden increase in the value of this measure may be due to the performance degradation of the disk, network congestion or a request on the disk that is taking too long to complete.
IO_queued Indicates the average number of I/O requests that are queued but are yet to be issued to this disk. Number  
Total_transfers Indicates the rate at which data transfer is being initiated from this disk. Transfers/Sec  
User_read_blocks Indicates the rate at which the blocks are read from this disk upon a user request. Blocks/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 disks to know which disks service block read requests slowly.
User_read_laten Indicates the average time taken to read a block from this disk upon a user request. Secs  
User_reads Indicates the rate at which the read requests from the user are serviced by this disk. Reads/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 the disks to know which disks service read requests slowly.
User_write_laten Indicates the average time taken to read a block from this disk upon a user request. Secs  
User_write_blocks Indicates the rate at which the blocks are written to this disk upon a user request. Blocks/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 disks to know which disks are servicing block write requests slowly.
User_write_laten Indicates the average time taken to write a block to this disk upon a user request. Secs  
User_writes Indicates the rate at which the write requests from the user are serviced in this disk. Writes/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 disks to know which disks are servicing write requests slowly.