4 7200 drive in raid0

[dale5605]

They do have a speed increase, anyone who benches on sandra will see that. Shoobie has no idea what he's talking about, maybe he accidentally set up a RAID 1 when he wanted to set up a RAID 0 or something

 

[HAVOC]

Well, i'm going to have to disagree with those that say that RAID 0 doesn't increase performance. I believe it does. And i've read many articles that agree with me. It isn't a HUGE noticable increase, but it is an increase none the less. RAID 0 writes data evenly across 2 identical drive. NO REDUNDANCY. Yes, if one drive fails they are both useless.

But, by writing data evenly over two drives, it has been proven in many tests that the read performance increases by as much as 12.5%. In the end, you might save 2-6 seconds in load times. For a serious gamer, that's enough for him to choose RAID 0. Also, RAID 0 arrays with a SEAGATE NCQ are the way to go.

If you want safety, go with a RAID 0+1 / RAID 1 / or RAID 5 "RAID 5 requires 3 or more drives, normal 4. If you have 4 (200GIG) HDD's in RAID 5 you actually only have 600GIG of HDD storage. A small portion of data is spread across 3 drives and the 4th drive stores parity of all those other 3 drives. IF one drive fails, you remove it, and plug and play a new SATA. Very expensive option though. Mostly used in Servers.

 

[Him]

Dude, you should do us all a favor and cite your sources. Not to prove you right or wrong, but to cease this argument permanently.

You also said:
"Well, i'm going to have to disagree with those that say that RAID 0 doesn't increase performance. I believe it does."

Kinda contradictory...

 

[ShoobieRat]

RAID-0 is more of a side-effect of RAID rather than a goal of RAID.

The true uses of RAID come in real-time backup for your data. Like in RAID-1, where you have two drives, but only one set of data. If one drive fails, the other was just a copy-cat so you haven't lost anything. RAID-1 is big for servers or workstations with really important data (or just anything you don't want blitzed).

RAID-0 is just one of those "hey, you can do this too" things. It's neat, it works, it's great if you need a lot of space for a project and you don't care about safety. If you want it for everyday use or for valuable stuff, RAID-0 is to be avoided.

 

[Chankama]

"Well, i'm going to have to disagree with those that say that RAID 0 doesn't increase performance. I believe it does."

Kinda contradictory...

Theoretically, it Raid 0 should be faster. Doesn't take much to think about why. And as for benches, there's lots on the Internet . The performance increase is not as flattering as I would've expected. But, it's still there. Once the algorithms that control the RAID IO accesses get smarter, I would think the performance of Raid 0 would increase. This is similar to what NCQ did to single hard drive accesses.

RAID-0 is just one of those "hey, you can do this too" things. It's neat, it works, it's great if you need a lot of space for a project and you don't care about safety. If you want it for everyday use or for valuable stuff, RAID-0 is to be avoided.

RAID 0 does decrease the "mean time to failure" of the whole HDD system. But, these days it's wise to make backups regardless of what ur setup is.

But, the point is RAID 0 is faster than nothing.

 

[HAVOC]

Dude, you should do us all a favor and cite your sources. Not to prove you right or wrong, but to cease this argument permanently.

You also said:
"Well, i'm going to have to disagree with those that say that RAID 0 doesn't increase performance. I believe it does."

Kinda contradictory...

How is my statement contradictory? Some people posted that it doesn't increase performance. I think at some level it does.

I actually just stumbled across a very indepth article that benchmarks various RAID arrays using different CPU's and Chipsets. The results actually are very interesting. With some chipsets & CPU's a RAID array did increase performance, with others the single drive out performed all raid arrays. So, i guess both sides of this arguement are right to some degree. Your performance in a single or RAID array will depend on your CPU and Chipset. Check this out. Many, Many pages and over 200 Benchmark Graphs...

http://techreport.com/reviews/2004q2/chipset-raid/index.x?pg=1

 

[dale5605]

That's a sweet article. I don't see where the performance dropped, realize the graphs are moved over because the numbers get larger. But the nforce3 chipset did well they don't even do the nforce4 chipset.
So the nforce4 is probably a whole lot better. That's why mine is so wicked fast.

 

[Apokalipse]

Raid 0 is not very much less reliable than a single drive
I mean, there are some pretty reliable drives out there

I would actually say a Raid 0 can in a lot of cases be less expensive than normal drives, considering capacity
if you want 400GB of storage, you can buy a 400GB drive for about $500 aus, or you can buy 2 200GB drives for about $170 aus each which makes $340

 

[Chankama]

Raid 0 is not very much less reliable than a single drive
I mean, there are some pretty reliable drives out there

Actually, it is. The Mean Time Before Failure of the system = MTBF of each HDD / # of HDD, assuming identical HDDs

 

[The51]

well how come wen i had one drive i was gettin transfer rates of 70mb/s and no i have 2 in raid0 i get 92mb/s?

Just want to point something out? If your running 2 SATA hard drives you should have a transfer rate of 150mb/s (roughly). Why are you saying your getting 92mb/s? That's not that fast...