There are new automatic latency threshold computations in SIOC. The latency thresholds for the SIOC can now be automatically set using the injector based models.
The benefit of this is that SIOC now figures out the best threshold for a datastore as opposed to using a default/user selection for latency threshold. The latency threshold is set to the value determined by the injector when 90% of the throughput value is achieved.
There is also a UI enhancement for this feature. SIOC is now turned on in stats only mode automatically.
It doesn’t enforce throttling but does gather statistics for helping SDRS. SDRS now has statistics in advance for new datastores being added to the datastore cluster.
Through device modeling, SIOC determines the peak throughput of the device. It then measures the peak latency value when the throughput is at its peak. The Threshold is then set (by default) to 90% of this value.
Admin still has the option to:
- Change % value.
- Manually set congestion threshold.
As we mentioned, in previous versions of Storage I/O Control, the default threshold value was set to 30ms. This wasn’t suitable for every different type of datastore. In vSphere 5.1, the new automatic threshold detection will use the I/O injector modelling to determine what the peak throughput of a datastore is. When peak throughput is measured, latency is also measured. The latency threshold value at which Storage I/O Control will kick in is then set to 90% of this peak value (by default). vSphere administrators can change this 90% to another % value or they can still input a millisecond value if they so wish.