Understanding the processing throughput reported by FlexRAID

When FlexRAID is computing parity, a processing throughput is shown in the status window.

The throughput presented by FlexRAID is a normalized value at the lowest denominator.
Essentially, FlexRAID low balls itself when reporting processing speed. 🙁

Under the normalized value approach, a system with 20 drives and 20TB of data will show the same processing speed as a system with say 3 drives and 3TB of data.
In reality, the throughput of the system with 20 drives would be far greater.

We were afraid that users would look at the much greater processing speed of large arrays and think there was something wrong with their system.
The way FlexRAID works, data is read and written in parallel. This means that a system with more drives can have its data processed in the same amount of time as a system with less drives. Put another way, FlexRAID can process 20TB of data over 20 drives for the same duration it would take to process 3TB of data over 3 drives.
The only limit to this is your system I/O (how much throughput your disk controller can sustain across multiple drives).

So, to find the real throughput for your system, multiply the throughput shown by the number of drives in your system.

Be Sociable, Share!

No comments yet.

Leave a Reply