Article Overview

For SafeHaven-4.0.1 and later released, there is a built-in monitoring tool shipped with the Windows SafeHaven agent. This document shows how to enable this tool to get statistics of the Windows disk IOs for a given period of time. It is a useful tool to have in order to determine the sufficient amount of WAN replication bandwidth for a given a workload, particularly for those busy servers that may be having problem meeting the recovery point objective (RPO).

Disk Monitoring

By default, the monitoring is turned off. To turn it on, simply run the DgSyncEx command with Administrator privileges

• first select the disk that needs to be monitored

DgSyncEx> select disk 0

• then enable statistics monitoring

DgSyncEx> set stats on

• then the stats command would report on the IO statistics from enabling monitoring to current time, for example,

DgSyncEx>stats

DgReportDiskUsageStats: Disk usage statistics for Disk #0:

"Collect disk usage statistics" is ON.

Start time: 13132276047620385 microseconds (13132276047620 miliseconds)
End time  : 13132276074343870 microseconds (13132276074343 miliseconds)
Difference: 26723485 microseconds (26723 miliseconds)

READ STATS:
  Count      : 86
  Sum        : 1129984 Bytes (1 MB)
  Average    : 0.04 MB/s

WRITE STATS:
  Count 512  : 18
  Count x512 : 49
  Count 4096 : 97
  Count x4096: 103
  Count All  : 152
  Sum 512    : 19470848 Bytes (18 MB)
  Sum 4096   : 515072 Bytes (0 MB)
  Sum All    : 19985920 Bytes (19 MB)
  Average    : 0.73 MB/s

FAILED WRITE STATS:
  Count      : 0
  Sum        : 0 Bytes (0 MB)

SYNC STATS:
  Count      : 0
  Sum        : 0 Bytes (0 MB)
  Average    : 0.00 MB/s

FAILED SYNC STATS:
  Count      : 0
  Sum        : 0 Bytes (0 MB)

DgSyncEx>

There is very little performance impact by turning on the monitoring feature. In order to turn it off, you can run

DgSyncEx> set stats off