dslreports logo
    All Forums Hot Topics Gallery
spc
Search Topic:
uniqs
1413
share rss forum feed


Octavean
Premium,MVM
join:2001-03-31
New York, NY
kudos:1

1 recommendation

SSD Raid 0 Array 4x60GB Setup Test Run

This was more of a test then anything else. I just wanted a basic idea of how it would feel. The system configuration is as follows:

Intel Core i5 2500K
Coolermaster Hyper 212 Plus
ASUS P8P67 Pro
8800GTS
4GB DDR3 1600 G.Skill (2x2GB)
Kingston SSDNow V series SNV425-S2 64GB (x3)
OCZ Agility 60GB (x1)
Corsair 750W PSU

And so on,…..

The RAID 0 array is setup on the four onboard Intel SATA II ports. Initially I didn’t want to mix dissimilar SSD units on the same RAID array. These SSD units were bought separately over time. Anyway, I’ve posted the HD Tune score for a single Kingston SSDNow V series SNV425-S2 64GB unit as well as a 2x64GB (128GB) RAID 0 array of the same SSD model. There is also a 3x64GB (192GB) RAID 0 array HD Tune score and finally a 4x60GB (240GB) RAID 0 array score (3x64GB + 1x60GB).


sk1939
Premium
join:2010-10-23
Mclean, VA
kudos:10
Must be very responsive.


Subaru
1-3-2-4
Premium
join:2001-05-31
Greenwich, CT
kudos:1
reply to Octavean
nice


norwegian
Premium
join:2005-02-15
Outback
kudos:1
reply to Octavean
What I can't understand is the linear speed. I gather platter are different and rightly so.

Down to the nitty gritty....what happened to the 4th drive?
All the drives match until drive 4?
--
The only thing necessary for the triumph of evil is for good men to do nothing - Edmund Burke



Krisnatharok
Caveat Emptor
Premium
join:2009-02-11
Earth Orbit
kudos:12
reply to Octavean
Any idea why the burst rate for 2 was much higher than 3 or 4?

Is that the full or trial version of HD Tune?


Pyrion
Liquid Metal Nanomorph

join:2001-12-01
Poway, CA
kudos:1
reply to norwegian
said by norwegian:

What I can't understand is the linear speed. I gather platter are different and rightly so.

Linear speed due to it being an array of solid state drives. No platters.
--
"There is much pleasure to be gained from useless knowledge." - Bertrand Russell


Octavean
Premium,MVM
join:2001-03-31
New York, NY
kudos:1

1 edit
reply to norwegian
said by norwegian:

What I can't understand is the linear speed. I gather platter are different and rightly so.

Down to the nitty gritty....what happened to the 4th drive?
All the drives match until drive 4?

I purchased these SSD units individually over time so I didn’t buy them all at once. Typically buying the model I was most comfortable with at a price point that seemed to be a bargain at the time. For example, the first Kingston SSDNow V series SNV425-S2 64GB unit I bought cost me something like ~$111 when they were typically going for ~$150 or more (way back when). The OCZ Agility model was a perceived bargain at the time as well and enough of a bargain for me to try something different. The Kingston units are fairly consistent performers though, even without TRIM in a RAID array:

quote:
Author: Allyn Malventano
Date: February 17, 2010
Subject: Storage
Manufacturer: Kingston

TRIM

I don't have and pretty charts or graphs to explain this next part, but I will share an observation I made during my fragmentation testing. When running my fragmentation tool, I observe IOPS drop as the drive becomes more and more overloaded with the task of tracking the random writes taking place. Here the JMicron controller behaved like all other drives, but where it differed is what happened after the test was stopped. While most other drives will stick at the lower IOPS value until either sequentially written, TRIMmed, or Secure Erased, the JMicron controller would take the soonest available idle time to quickly and aggressively perform internal garbage collection. I could stop my tool, give the drive a minute or so to catch its breath. Upon restarting the tool, this drive would start right back up at it's pre-fragmented IOPS value.

Because of this super-fast IOPS restoring action, and along with the negligible drop in sequential transfer speeds from a 'clean' to 'dirty' drive, it was impossible to evaluate if this drive properly implemented ATA TRIM. Don't take this as a bad thing, as any drive that can bring itself back to full speed without TRIM is fine by me, even if that 'full speed performance' is not the greatest.

This type of self-healing (i.e. without needing TRIM) is great for those wanting to run a few SSD's behind a RAID, since no RAID implementation is currently capable of passing TRIM from the OS to the arrayed SSD's. Better yet, considering this drive is tailored to the budget crowd who may very well still be running XP or Vista, it's good to have a few choices that don't require TRIM to maintain decent levels or performance.

»www.pcper.com/reviews/Storage/Ki ··· rance/Fr

said by Krisnatharok:

Any idea why the burst rate for 2 was much higher than 3 or 4?

Is that the full or trial version of HD Tune?

I always thought that the burst rate was in error and simply ignored it. I was using the free version of HD Tune.

For what its worth,...

I’ll add that I suspect that the throughput limits of the Intel SATA II controller are being hit here with four SSD units in a RAID 0 array. I can’t really be sure until perhaps I have an Intel X79 motherboard with four Intel SATA 6G ports (RAID) or a good RAID card. The ASUS P8P67 Pro motherboard has four SATA 6G ports but two are Intel and two are Marvel.


rusdi
American V
Premium,MVM
join:2001-04-28
Flippin, AR
kudos:2
said by Octavean:

said by norwegian:

What I can't understand is the linear speed. I gather platter are different and rightly so.

Down to the nitty gritty....what happened to the 4th drive?
All the drives match until drive 4?

I purchased these SSD units individually over time so I didn’t buy them all at once. Typically buying the model I was most comfortable with at a price point that seemed to be a bargain at the time. For example, the first Kingston SSDNow V series SNV425-S2 64GB unit I bought cost me something like ~$111 when they were typically going for ~$150 or more (way back when). The OCZ Agility model was a perceived bargain at the time as well and enough of a bargain for me to try something different. The Kingston units are fairly consistent performers though, even without TRIM in a RAID array:

quote:
Author: Allyn Malventano
Date: February 17, 2010
Subject: Storage
Manufacturer: Kingston

TRIM

I don't have and pretty charts or graphs to explain this next part, but I will share an observation I made during my fragmentation testing. When running my fragmentation tool, I observe IOPS drop as the drive becomes more and more overloaded with the task of tracking the random writes taking place. Here the JMicron controller behaved like all other drives, but where it differed is what happened after the test was stopped. While most other drives will stick at the lower IOPS value until either sequentially written, TRIMmed, or Secure Erased, the JMicron controller would take the soonest available idle time to quickly and aggressively perform internal garbage collection. I could stop my tool, give the drive a minute or so to catch its breath. Upon restarting the tool, this drive would start right back up at it's pre-fragmented IOPS value.

Because of this super-fast IOPS restoring action, and along with the negligible drop in sequential transfer speeds from a 'clean' to 'dirty' drive, it was impossible to evaluate if this drive properly implemented ATA TRIM. Don't take this as a bad thing, as any drive that can bring itself back to full speed without TRIM is fine by me, even if that 'full speed performance' is not the greatest.

This type of self-healing (i.e. without needing TRIM) is great for those wanting to run a few SSD's behind a RAID, since no RAID implementation is currently capable of passing TRIM from the OS to the arrayed SSD's. Better yet, considering this drive is tailored to the budget crowd who may very well still be running XP or Vista, it's good to have a few choices that don't require TRIM to maintain decent levels or performance.

»www.pcper.com/reviews/Storage/Ki ··· rance/Fr

said by Krisnatharok:

Any idea why the burst rate for 2 was much higher than 3 or 4?

Is that the full or trial version of HD Tune?

I always thought that the burst rate was in error and simply ignored it. I was using the free version of HD Tune.

For what its worth,...

I’ll add that I suspect that the throughput limits of the Intel SATA II controller are being hit here with four SSD units in a RAID 0 array. I can’t really be sure until perhaps I have an Intel X79 motherboard with four Intel SATA 6G ports (RAID) or a good RAID card. The ASUS P8P67 Pro motherboard has four SATA 6G ports but two are Intel and two are Marvel.

That's what I see. Really no bottleneck, just the limits of the SATA II interface. I'd say you pretty much *maxxed out* that pipeline!!

Very impressive, yet I am kinda' wondering if there would be any noticeable speed, (perceived or real) in day-to-day operation?
I mean, can you really tell a difference, when at that level of IOPS?
Pfft! who needs SATA III?
--
Constitutional law. Government out of my business. Strong/stable Dollar. These are a few of my favorite things.


Octavean
Premium,MVM
join:2001-03-31
New York, NY
kudos:1

1 recommendation

Actually no, I don't really notice any speed difference. Good point though.

The Kingston SSDNow V series SNV425-S2 64GB and OCZ Agility (v1) are not the fastest SSD models even for their time. However, they are fast enough to let you know that you are on an SSD not a HDD. So basically fast is fast. Its like the difference between 100 fps and 200 fps in a first person shooter, your eyes and brain can't pickup on the extra 100 fps but they are there anyway.

Also, I don't think HD Tune is a very good benchmark for this. Sequential reads / writes are only part of the story. I suspect random reads / writes are more indicative of normal everyday use,...

but then again not really sure. I'm not really into benchmarking and I don't think I am that knowledgeable about this subject.


rusdi
American V
Premium,MVM
join:2001-04-28
Flippin, AR
kudos:2
I can see a scenario when that kinda speed is desirable, and advantageous. Perhaps writing programs when extremely large chunks of data are compiled, or when editing large video files, etc.

Now I'm reading about "bonded SATA III, SSD (with RAID interface "on controller"). This looks to be "the next generation" of SSD. One can only imagine what kinda transfer speeds something like this can perform.

I guess the quest, (and need) for speed will never end! This stuff, simply amazes me!!!

--
Constitutional law. Government out of my business. Strong/stable Dollar. These are a few of my favorite things.


pnjunction
Teksavvy Extreme
Premium
join:2008-01-24
Toronto, ON
kudos:1
reply to Octavean
said by Octavean:

I just wanted a basic idea of how it would feel.

Really is 700-900 MB/s going to feel much different than 500 MB/s?

Admit it you worked up the biggest boner you could and and reached for the ruler. Don't worry we've all done it.