Samsung F3 500GB's in RAID0 200GB Array, quick benchies

So I grabbed a couple of these to replace the single, VERY slow 400GB drive I was using previously as my system drive. I was too impatient to install an OS and run benchies on a single drive, full drive, etc. But I can at least give people an idea of how these run in my setup.

Q9550 - Lapped - 4012Mhz
Gigabyte GA-P45-UD3P
2x 2GB Gskill DDR2-1000
2 x 500GB Samsung F3 - RAID 0



I have a 200GB RAID 0 array for the system and then the rest RAIDed for extra storage. I know this doesn't really give you guys a perfect idea of how a single drive performs, but it's all I have







I'm quite happy with how fast the system feels. I can finally max out my 1TB on transfers. I also have my old Seagate .11 1.5TB drive in an eSATA enclosure and it topped out at over 100MB/s, which is nice (I'm used to 30-40 MAX)

ummm those result are interesting for 2 drives...

Why is it that hardly anyone runs iops/random 4k write tests with hdd's but it's practically a requirement with ssd's?

SuperMiguel said:

ummm those result are interesting for 2 drives...

Click to expand...

Is that good or bad?

Know Nuttin said:

Why is it that hardly anyone runs iops/random 4k write tests with hdd's but it's practically a requirement with ssd's?

Click to expand...

How would I go about doing that (if that is what you are asking for).


I don't really know a ton about HD benchmarks, for me, I was just looking at my average transfers, etc. I don't entirely know what I'm looking at (That's kind of why I posted, to get others opinions )

I'd say the results seem higher than they should be.

IOmeter to test for 100% random writes, 4k in size, 8GB in size, outstanding I/O's per target of 3 is what the standard seems to be for testing SSD's. Here is how to configure it

http://mydellmini.com/forum/dell-mi...ter-benchmarks-latency-stutter.html#post43019

You could also do a CrystalDiskmark test though I don't find it as informative. Samsung 2nd gen SSD's do well in terms of MB/sec but IOPs aren't represented doing CDM, and that is where it is not so good.

The benchmarks you used are mainly sequential tests which is great assuming your data is all perfectly in line and the drive can just read them without having to move around to different areas of the disk. It doesn't take into account much of the way Windows does writes or perhaps the way data is really aligned on your system.

Know Nuttin said:

I'd say the results seem higher than they should be.

Click to expand...

+1

Know Nuttin said:

http://mydellmini.com/forum/dell-mi...ter-benchmarks-latency-stutter.html#post43019

Click to expand...

drizek said:

Either make sure that you have only 310mb of free space on the drive or figure out how many sectors add up to 310mb, and enter that number under maximum disk size

Click to expand...

So... a sector = 512 bytes.. So 310 MB = 325058560 byes.. Meaning that 310 MB = 634880 sectors??

So we enter 634880 under maximum disk size?

SuperMiguel said:

+1

Click to expand...

I'd guess that's a good thing then, no?

jivetrky said:

I'd guess that's a good thing then, no?

Click to expand...

i guess

Know Nuttin said:

The benchmarks you used are mainly sequential tests which is great assuming your data is all perfectly in line and the drive can just read them without having to move around to different areas of the disk. It doesn't take into account much of the way Windows does writes or perhaps the way data is really aligned on your system.

Click to expand...

I'm encoding some video right now...but later on I'll try the IOMeter

My thoughts though are that I ran these because these are the ones that I see people using most. I just wanted to get an idea where they stand against other drives to see how these compare.

jivetrky said:

I'm encoding some video right now...but later on I'll try the IOMeter

My thoughts though are that I ran these because these are the ones that I see people using most. I just wanted to get an idea where they stand against other drives to see how these compare.

Click to expand...

Don't get me wrong, and don't let me ruin your enthusiasm for your new drives. Not my intention at all. Enjoy them if they make you happy, and to hell with anyone who thinks they aren't any good.

I was just curious to see what a RAID 0 array does on the same tests as SSD's seem to be measured by.