For most applications there are two basic usages that the user can access with this controller.
The first is just the typical RAID. In this version, the controller and its driver do the ‘grunt’ work of RAID processing. This utilizes the onboard processor more as it has to launch the device and load the operating system. Once the operating system is loaded, the majority of the RAID processing is handed off to the host computer. This mode presents the sum of the drives to the operating system as one volume.
Next we have ‘Soft’ (software) RAID. In this type of application the user can configure the devices that are connected to the controller in a pass-through mode as JBOD disks. Essentially, this allows the user to present the disks to the operating system as individual devices. This allows the device to merely pass the Input/Output operations through to the operating system with no computational overhead from the device itself. Pass through is also a feature typically used by individuals who are looking to provide 6Gb/s speeds on previous generation hardware.
The most commonly used Soft RAID is operating system based dynamic disks. This allows the user to leverage more of the power of the CPU in their system to handle the RAID overhead. This can result in decidedly faster performance with the notable drawback of the volume not being able to boot.
This Soft RAID implementation can be commonly used in several different home/file server applications so the results are relevant to our testing. Perhaps more importantly the purpose of the Software RAID will highlight the performance of the controller when it is merely a JBOD (Just a Bunch Of Disks) configuration, also referred to as pass-through. This will be the main use of this product for those looking to upgrade existing motherboards.
We will be highlighting the results of both scenarios in RAID 0 for our testing of this controller.
RAID 0 TESTING “ AS SSD
AS SSD is a forerunner of SSD testing, as it covers many important metrics relating to SSD performance. There are multiple testing tools that can be used, and here we will use them to highlight the ability of the device to handle SSDs, even though it clearly isn’t designed with SSDs solely in mind. For such a budget minded solution it does manage to impress with the sequential access. The random read is a bit lacking at the higher Queue Depths.
There is definitely a notable step down in performance when the controller is handling the load.
The .ISO testing suite within the AS SSD benchmark shows some usage patterns that are indicative of certain copy usage patterns. Different file types will handle differently, and this test highlights those differences. With this type of usage, the differences in speed amongst the two different approaches are not quite as large.
Finally the Compression Suite below rounds out the AS SSD testing suite. This battery of tests emphasizes the differences in speeds with different levels of compressible data. The C400/M4 handles data very evenly and does not rely on compression as much as other solutions. Again these tests show a bit of the difference between the two types of RAID.