Adaptec RAID 2405 SAS / SATA Controller

WinAVI, Peazip & Dummy File Creation Results

WinAVI Video Encoding
The transcoding of video files between different formats for use on devices such as portable media players or burning to DVD is becoming increasingly common in today's digital world. While the method of conversion is largely CPU intensive, it requires a fast hard disk to keep up with the stream of data. For this stage of the testing we used a 600mb Xvid encoded AVI file placed on each of the hard disks and then used WinAVI to transcode the file into DVD format. The time taken was recorded with a stop watch.
Once again the Adaptec + Seagate (Array1) configuration loses out to the Intel + Western Digital array mainly due to the slower speed of the Seagate Barracuda Drives. However, as soon as the Adaptec controller is teamed up with the Western Digital Velociraptors the tables turn and Array3 becomes the fastest of the bunch. However, with only 3 seconds separating the slowest and fastest arrays it could possibly show that the encoding of our AVI file was more bottlenecked by CPU performance than hard disk speed.
Peazip File Compression
File compression is yet another area where the system is just as reliant on the performance of the CPU as it is on the hard drive. To simulate the compression of various types of files, a folder containing a collection of 200 text documents filled with random contents and file sizes varying from 1KB to 100MB was copied to each of the arrays. This folder was then compressed and decompressed using a utility called Peazip which provided us with an accurate "time taken" reading.
With less than a second between each of the Compression and Decompression results it looks like CPU performance may well be a bottleneck here as well. However, the combination of Adaptec 2405 controller + WD Velociraptor hard disks is at the front of the pack, showing that although the difference is small, there most certainly is one.
Dummy File Creation
When performing manual "file copy" benchmarks, the performance of the drive that the files are being copied from can directly and negatively affect the results of the drive they are being copied to. This is something that needs to be taken into consideration when benchmarking high performance RAID arrays as the performance is easily in excess to that of a standard hard disk. Therefore, to test the write performance of each array, a freeware utility called Dummy File Creator was used to generate files directly to each of the hard disks removing the possibility of any bottlenecks.
The results below show how long each of the drives took to write 20GB of files with sizes ranging from 1MB to 1GB.
In contrast to the previous two benchmarks, Dummy File Creation uses up very little CPU resources, instead concentrating on the array and filesystem write speed. Despite this, there is still very little difference between the Intel Matrix based RAID array (Array2) and Adaptec Array3 indicating that although our HDTune results put the Adaptec controller ahead of the Intel Matrix Firmware based array, there really is very little real world difference.
«Prev 1 2 3 4 5 6 7 Next»

Most Recent Comments

10-11-2008, 07:29:56

"With an 800MHz processor, 128MB of DDR2 and a PCI-E 8x interface, can Adaptec's latest entry-level SAS/SATA RAID controller offer a boost in performance?"

Adaptec RAID 2405 SAS / SATA ControllerQuote

10-11-2008, 11:26:39

Nice one Jim!

IT would seem that for a fairly small raid 0 array it is not really worth it...

It would be nice to see a 4 or even 6 disk comparison sometime in the future Quote

10-11-2008, 13:43:29

I wondering ...

Why you don't mention missing RAID 5 (6) as "BAD".

My opinion, it is very bad.

The second thing that was more then surprising is 2 HDD configuration.

Dedicated Hardware RAID controllers may perform on the same level (or even worse) as software assisted with 1 disk attached.

With 4 disks you can get 300+ MBps read write.

Once, during my work in HP IT department i had an opportunity to play with few RAID setups and found that dedicated hardware controllers begin significantly outperform software assisted controllers like onboard controllers on almost any M/B when used in RAID 5 or 10 setups in system under load.

It worse nothing to compare them like you did.

Try to ran on top of your OS DB + Web serving 500+ connections per minute at least + ran some serious CPU intensive task (like C or C++ compilation) and then you'll see what you paid for with Hardware RAID controller.

But the bottom line, It is good to see that this site reviews not only i7 or HD4870 like others do.

Keep doing it Quote

Register for the OC3D Newsletter

Subscribing to the OC3D newsletter will keep you up-to-date on the latest technology reviews, competitions and goings-on at Overclock3D. We won't share your email address with ANYONE, and we will only email you with updates on site news, reviews, and competitions and you can unsubscribe easily at any time.

Simply enter your name and email address into the box below and be sure to click on the links in the confirmation emails that will arrive in your e-mail shortly after to complete the registration.

If you run into any problems, just drop us a message on the forums.