• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

About RAID-0 Array ?

I

isaac27

Limp Gawd
Joined
Feb 11, 2007
Messages
152
I have a question about RAID-0
I HAVE this ASUS P5W DH it have RAID CONTROL BUILT-in

My question : is there a big difference when you write to a single SATA HD
or when you write to- 2 HD in RAID-0 My Teacher (MCSE) Say
"that the data is split for 2 when you write into 2 Hard-Drive in RAID 0
for exc. - if you copy 700mb file then the time that will take to write it on
RAID-0 HD will be the time that will take you to write 350mb"

is there someone who try to use Raid-0 After using A single HD Mode.....
Can you Please tell me what is the real difference
I value your advice into my account

Thanks
Isaac
 
You WILL see a difference when you go from a single drive to 2 drives in RAID 0. I don't think it'll cut down your write performance to 50% like your teacher says it will. In theory, it will copy 350MB to one drive and 350MB to another. However, there is still some overhead using a RAID array that prevents you from getting that 50% reduction. I think the general consensus is that you'll notice a 20-30% increase in performance.
 
If I remember right, the P5W's southbridge is an ICH8, which handles Raid array's really well from the benchmarks I've seen. I run a Raid 0 on a 680i and I can tell you its really nice. Though your average seek times do go up its only in mili-seconds where as your Read rate goes up in hundreds of megabytes. So yeah, there is a big diffrence when compared to writing to a single SATA - a Raid 0 is even faster than a Raptor drive.
 
Cyrilix said:
http://www.pcguide.com/ref/hdd/perf/raid/levels/singleLevel0-c.html

RAID 0 is for performance. If you google up some benchmarks, you will also see its effect on performance.
Click to expand...

<grr> I had a large post written up for FF crashed and it's gone. Since I am not going to try and reproduce that, let me be brief: I suggest that the OP searches this subforum for the following phrase "RAID-0 +gaming" which will yields tens of threads that contain information about the subject. It will explain the strengths of RAID-0 (large continuous reads and writes) and the discussion on access patterns that games and other applications use.

J_Devil said:
...a Raid 0 is even faster than a Raptor drive.
Click to expand...
It depends how you define `fast'. Surely a dragster with 680KW is going to be faster in a straight line than a Porsche 911 with ~200KW, but when it comes to driving around a racetrack -and I mean a real one, not the oval that Nascar uses- the Prosche will certainly be faster than the dragster.
 
J_Devil said:
If I remember right, the P5W's southbridge is an ICH8, which handles Raid array's really well from the benchmarks I've seen. I run a Raid 0 on a 680i and I can tell you its really nice. Though your average seek times do go up its only in mili-seconds where as your Read rate goes up in hundreds of megabytes. So yeah, there is a big diffrence when compared to writing to a single SATA - a Raid 0 is even faster than a Raptor drive.
Click to expand...

The P5W DH uses the ICH7R. It also has a Silicon Image 4723 controller and a JMB363 controller for RAID purposes. The interesting thing is the Silicon Image 4723. It is supposedly a full on hardware RAID rather than a Hybrid software/hardware solution like what's on the motherboard. As far as the Intel ICHxR series, they are faster than most onboard RAID controllers including the NVRAID controllers.

On another note I don't think that a RAID 0 will always be faster than the Raptors. Raptors have higher rotational speed for their platters and with fewer platters being used in the Raptors than there are in other drives, and with pretty good platter density, they have really low seek times. I'd say that it would depend on the situation as to wether or not a particular RAID array would be faster than the Raptor in any given circumstance.
 
J_Devil said:
Though your average seek times do go up its only in mili-seconds where as your Read rate goes up in hundreds of megabytes.
Click to expand...

I don't even know where to begin with this. Milliseconds matter, a whole lot. Drives can only deliver new data while they're not seeking (yes, they could pull out of cache, but I don't think that happens a lot) and a millisecond is a long, long time when compared to the CPU that can do a million instructions (!) in that time. Adding even a little bit of latency puts a lot of hurt on a disk system, no matter how much STR you get to make up for it. Disks are dog-slow, and adding latency to them is something to avoid in all cases.

What do the units have to do with it? If you express STRs in gigabytes per second, and seek times in nanoseconds, would that make one more important? My STR is around 0.09 gigabytes per second, after all, and my seek times are in the neighborhood of 8,300,000 nanoseconds! These numbers are unrelated, and the units have nothing to do with each other.

As drizzt81 said above, the artificial benchmarks people run don't necessarily have anything at all to do with what they plan to do with the drive in real usage. Would you buy a car based on its straight-line performance, or might you want to try driving it in normal usage for a week? I'd like to test my drives with the applications I plan to run on them in real usage, not just buy whichever does the best in HDTach, and I suggest you do the same.
 
unhappy_mage said:
As drizzt81 said above, the artificial benchmarks people run don't necessarily have anything at all to do with what they plan to do with the drive in real usage. Would you buy a car based on its straight-line performance, or might you want to try driving it in normal usage for a week? I'd like to test my drives with the applications I plan to run on them in real usage, not just buy whichever does the best in HDTach, and I suggest you do the same.
Click to expand...

Yeah I used to have a better analogy:
``Basing a HDD purchase on a HDTach benchmark result is like buying a Video card based on the "theoretical pixel fill rate" or similar metric."
 
You must log in or register to reply here.
Back
Top