Symptom: Very poor throughput when writing files to the SnapServer
This issue can have multiple causes ranging from server load to network bottlenecks, poor packet management, bad cabling, misconfigured switches etc. Systems that are resyncing after a failed disk is replaced will also have reduced throughput until the sync is complete.
If you have verified that the above possible scenarios are not at issue another possibility is the write cached volume option.
New for GuardianOS 6.0 this option disables or enables write caching at a hardware level. By default it is enabled to enhance performance (In previous versions of GuardianOS it was enabled by default and it is not possible to disable).
You may wish to disable the volume write cache to enhance data preservation for business critical applications, however as this option causes the Snapserver to immediately flush and sync all writes to disk it does come at the cost of a very significant performance hit. Disabling this option is not recommended unless you understand this impact and require the enhanced protection.