LSI RAID Controller help, 2308 chip on-board ASRock Extreme 11 motherboard is slow!

[Callsign_Vega]

Request some help for those who know about the LSI controllers. Using the 2308 controller chip found on the ASRock Extreme 11 motherboard and I have eight 128 GB OCZ Vertex 4 SSD's (Firmware 1.5) set up in RAID 0.

This is the very poor speed I am getting:




This is the speed I got with a mere two 128 GB Vertex 4's on Z77 Intel Raid 0:





Both are 64kb stripe.

Pretty much the only speed that looks correct on the LSI setup is the 4k read and Acc.time's. Everything else look's very slow. The 4k write is abysmal, and the sequential speed is far far below what eight M4's got here using the same controller chip (4180 MB/s seq read):

http://thessdreview.com/our-reviews/...ick-preview/5/


These are the settings:













Now in that last screenshot above, it shows disk cache enabled (which doesn't effect the speed with before and after change runs), read policy to none, write policy: write through.

But when I go to change any of those items by right clicking the virtual drive properties, you can see the window, the only option is disk cache policy.

According to this site: http://kb.lsi.com/KnowledgebaseArticle16553.aspx

There are a lot more options there that I don't have. What gives?

Anyone have a clue why this performance is so poor?

Am I suppose to buy some of this software for hundreds and hundreds of dollars located here: http://store.lsi.com/store.cfm/Advan...tware_Options/

Just to get my LSI controller speed working properly?

[johnw]

Have you measured with anything other than AS-SSD?

Why not try IOMeter like the article you referenced?

[Computurd]

that looks to be uber sauce on that extreme11. of course Asrock has to get it right. the reason you cant do all of the options is that MSM covers all raid controllers, even with standard HBA you cannot access all options, unless they pertain to your card.

I recently did a spin with the 9207, which has the same ROC and pulled 633,000 IOPS fairly easily. something is definitely amiss.

[Callsign_Vega]

Below are screen shots of how ASRock implemented this chip and how I configured it in RAID.






The BIOS of the ASRock motherboard is P1.10.

I take it the configuration utility in my screen shots is HBA and not WebBIOS?

I noticed under the MegaRAID utility it doesn't allow me to change the write policy and keeps it write-through. Is there a way to force that to write back? (I have a UPS for the computer, but the motherboard LSI chip does not have a battery backup). I only have one option (Disk Cache Policy) and not all of the options as found here:

http://kb.lsi.com/KnowledgebaseArticle16553.aspx

Under MegaRAID my settings are: Access Policy- Read/Write, Disk Cache Policy- Enabled (only settings I can change), Read Policy- No read ahead, IO Policy- Direct IO, Write Policy- Write Through.

Also, in the BIOS configuration utility there is absolutely nowhere to change the stripe size during volume create. Is this on purpose? It only uses the default 64kb. Some say that my low performance may be because I am not using 128kb stripe, but I highly doubt that as 64kb has done really well in mixed-use scenarios in recent RAID tests versus other stripe sizes.


Ok, so I re-flashed the firmware of all V4's to 1.5 and secure erased. I set two of them up on Intel X79 RAID 0 with 64k stripe and tested 2x, 4x, and 6x V4's on the LSI 2308 RAID 0 64k stripe (only one possible apparently).

For a refresher, this is 2x V4's on Z77 with write-back on:




This is 2x V4 on Intel X79 RAID 0 with write-back on:



As you can see, Z77 always come in quite a bit faster than X79 in RAID 0 for some reason. I've seen the same thing over many different motherboards. Anyone know why?


This is 2x V4's on X79 with write-back off:



For these results, it seems like the only major difference between having write-back on and off is the 4k write speed.


Now all LSI tests below have write-back off seeing as you cannot enable it. 2x V4's on LSI 2308:



Pretty much a drop across the board versus X79 besides 4k read/write.


4x V4 on LSI:



Here seq write doubles, but seq read only increases about 80% for some reason. 4k write actually decreases and I don't know how, but 4k-64Third stays the same.


6x LSI:



Here 6x drives is around 280% the speed of 2x drives in seq write, but drops down to only around 220% of 2x drives in seq read (300% being perfect of course). No change in 4k-64third versus 2x and 4x really. The only difference with 6x is somewhat lackluster seq speed increase.


And the original 8x LSI test that got me questioning all this:



Seq write is doing pretty good, 367% out of a perfect 400% versus two drives. Seq read is much worse, only 290% versus two drives. Not sure why it's so low when running 8x PCI-E 3.0.

Once again 4k-64Third is pretty much the same as 2x drives. Measureably lower than X79 and Z77 with only two drives. I knew with RAID 0 the 4k read/write speeds would not increase, but I surely thought 4k-64Third would.

So basically, there is no need for this LSI controller to be on PCI-E 3.0 8x as it doesn't come close to saturating PCI-E 2.0 8x. Not sure how ASRock claim 3.8 GB/s using this controller.

All I can think of is maybe crappy firmware on the 2308 chip, some sort of bad drivers or this LSI chip is just not that good/too stripped down (no memory cache) and being integrated to really shine.

Here is a user with the exact 8x Vertex 128GB drives but using a LSI 9265-8i RAID card:



Seq are about identical, but his 4k write and 4k-64Third speed destroys mine.

Any thoughts, anything I can test while I have the OS on 2x V4's X79 and 6x V4's on LSI? If this is as fast as it's going to get, for $1000 in drives and $250+ for the LSI controller I may think about reverting to something else. And apparently Fastpath won't work on this chip, so that rules out any help there.

Hopefully the LSI tech can shed some light on the topic.

[Callsign_Vega]

I just tested with IOMeter the same settings as in that Hexus video at Computex in which ASRock was getting 3500 MB/sec 1MB seq Read. I can only get 2100 MB/s with the same settings in IOMeter. Not sure how they've gotten that fast of speed.

[Andrea deluxe]

I just tested with IOMeter the same settings as in that Hexus video at Computex in which ASRock was getting 3500 MB/sec 1MB seq Read. I can only get 2100 MB/s with the same settings in IOMeter. Not sure how they've gotten that fast of speed.

Start aida64 and tell me the speed of connection of lsi chipset!

For me is a innterconnection bottleneck

Is too many drive on a single chip connected mybe @4x pciexpress

[Computurd]

configure your disks as separate volumes, then access them individually with a worker apeice, QD32 per worker. You will pull some numbers. can you link the Hexus video? i would like to see it

[Callsign_Vega]

Start aida64 and tell me the speed of connection of lsi chipset!

For me is a innterconnection bottleneck

Is too many drive on a single chip connected mybe @4x pciexpress

configure your disks as separate volumes, then access them individually with a worker apeice, QD32 per worker. You will pull some numbers. can you link the Hexus video? i would like to see it

I will try the individual volumes worker test.

Here they have a video of it hitting those speeds. http://hexus.net/tv/show/2012/06/ASR...11_motherboard

Something is very fishy here and I don't like it.

Ya, wait a second. How are they getting 3500+ MB/sec seq read's with eight ADATA SX900 SSD's? As far as I can tell those are far inferior to Vertex 4's and I am only getting 2100 MB/seq.

Something can't be right!

Got done talking with LSI tech support. According to them the 2308 chip is one of their newest chips and it is no slouch. Without external RAM it can do 600,000 IO's/second and should not be running nearly this slow. They included this link:

http://www.lsi.com/downloads/Public/...ct%20brief.pdf

Quote from LSI tech: "I know Vertex and LSI had some issues in earlier stages and I do believe the engineers from both LSI and Vertex have worked on performance as best as they can."

That doesn't sound too encouraging... I may have to order a couple of M4's to test them out and see if it is indeed the LSI RAID chip not working properly, or some conflict between the Vertex 4 and the LSI chip not playing nice as originally suspected.


What are your guys thoughts on all of this? Do you think this RAID setup is suppose to be this slow and it's operating correctly?

[Callsign_Vega]

6x LSI Hardware RAID and bootable:




6x JBOD Win 7 Stripe volume:




Why would the hardware bootable RAID option be so much slower? Seem's silly to have to run some drives off of the X79 intel ports to Run Win 7 and then only save programs to the LSI Array.

[Andrea deluxe]

Right!

Ocz fail another time!

Why you bought the vertex4??

M4 is always the best ssd

[Computurd]

it has PLX bridge chips. interesting. two of them, which means there are probably not any 'pure' slots on there. im sure that they are using the pcie lanes for the LSI as well.
don't underestimate them Adata SX900's, they can be uber in FOB btw.

for even higher speeds...do not RAID them with winders RAID. it is no good for scaling. configure them as separate volumes, then assign either a worker, or manager, to each. Then you will see top speed. no raid!

they have actually refined the 2308 and FW to where they can pull 700K IOPS now, i reached around 633K and my hardware was the limitation.

[Callsign_Vega]

Right!

Ocz fail another time!

Why you bought the vertex4??

M4 is always the best ssd

How so? Vertex 4 beats the M4 in almost every test:

http://www.xbitlabs.com/articles/sto...b-256gb_3.html

it has PLX bridge chips. interesting. two of them, which means there are probably not any 'pure' slots on there. im sure that they are using the pcie lanes for the LSI as well.
don't underestimate them Adata SX900's, they can be uber in FOB btw.

for even higher speeds...do not RAID them with winders RAID. it is no good for scaling. configure them as separate volumes, then assign either a worker, or manager, to each. Then you will see top speed. no raid!

they have actually refined the 2308 and FW to where they can pull 700K IOPS now, i reached around 633K and my hardware was the limitation.

The two PLX chip's are feed their own 16x lane from the CPU, one to each PLX which each has two 16x speed slots for GPU's. The LSI chip has it's own and the last 8x lane from the CPU, it does not go through the PLX chip(s).

So the 2308 is a pretty good chip, even with zero cache? People on other forums keep telling me the chip is no good and that it's a 'bargain basement basically software" raid chip and my slow speeds are because of it. Thoughts?

[mobilenvidia]

The SAS2308 is a faster clocked SAS2008, the SAS2308 supports PCIe Gen3, both are zero cache controllers.
In IT mode (passthrough) these are the best for ZFS and other OS's that do their own RAID.
In IR mode these controllers are as good as it gets with simple RAID (0,1 and 10)

Zero cache = you don't get write back, read ahead, Cached IO etc etc.
The only option you have is turning the drive cache on/off, with SSD's turn this off for better results.

Now the Extreme11 board has 2x PEX8747 PCIe bridges 16 lanes in and 32 lanes out each, these 64 lanes head of to the 4x PCIe 16x slots
Now the LGA2011 CPU has 40 PCIe Gen3 lanes, 32 go to the 2x PEX8747's, leaving 8 Gen3 lanes spare.

Now I can't be certain if all the 8 Gen3 lanes go to the SAS2308 or if other devices share lanes with it.
There are also 8x Gen2 lanes from the PCH, not sure how these are used either.
I do hope that the SAS2308 gets the 8x Gen3 lanes, even 8x Gen2 would be OK.
But if it only get 4x lanes then you will bottle neck as above.
There is a Datasheet on the Extreme11 board layout somewhere which has it drawn, can no longer find it.

LSI has the new P14 Firmware for the LSI9207 SAS2308 controller.
You should be able to update to this, as the onboard BIOS look the same as LSI's.
You will need IR firmware to have RAID.
You can upgrade to this via MSM in windows, load the Firmware then load the BIOS

Turn Diskcache OFF! in MSM, don't touch the Device manager Drive settings, default is king.
REBOOT for setting to take.
To help diagnose run a single drive test first, this is your base speed.
Setup a RAID0 array with 2 drives, REBOOT, get performance figures for this.
Keep adding a drive and do more tests.
At a certain point you may bottleneck, theoretically each drive you add should add the single drive performance.
Theoretically you should be able to get 8x ~500MB/s = 4TB/s

If the single drive speed is already slow then something is afoot, adding more drives ramp as slow.

When you get the speed sorted you'll be able to shift this to the eXtreme section

Would love to do a review on this powerhouse.

[Computurd]

it will not go as high in RAID, simple as that. if you were to configure the device as 8 individual volumes, then assign a worker to each, you will not be able to saturate that controller. as said above, this is intended for OS's that do their own RAID, and the windows RAID is horrible. that is a limitation of windows, not the hardware.
I can tell you that this is the setup that they were using with the video, 8 different volumes.
I have been able to achieve similar results, and my review is in the queue at TT to be out soon. IR just doesn't have that much sauce in it, but it is more for mirroring than anything.
bear in mind, this is an HBA, and a HBA chipset. you are expecting more than it will give you in RAID. Many HBAs dont even have integrated RAID.

people on other forums telling you that it is a bargain basement raid chip..... well its an HBA, what do you expect?
saying the chip is no good is inane. It is the highest performing chip on the market in its class, period. in passthrough its IOPS performance surpasses that of every HBA and/or RAID controller out there. how is that "no good?"

From reading at the [H] forum thread as well a few things....

You need to delete the firmware from the 2308 before you can switch to IT firmware. that is why it says you " Cannot Flash IT Firmware over IR Firmware!"

second..a few others have told you the key as well, configure as separate volumes, but yoiu have not tried. You have to understand that IR RAID basically 'boosts' the RAID into the OS, where some of the work is transfered over to the OS via the HBA driver. IE....it will suck just as bad as windows RAID.
if you want the numbers do as the review that you have read states, and many have suggested as well....separate volumes!

[Callsign_Vega]

So I put Win 7 on a WD 10K Raptor on the Intel X79 port so that I can freely change the 8x Vertex 4 LSI RAID setup.

This is what I get when using the LSI hardware RAID with boot-up config utiliity (not selected as a boot device and no Windows installed):



Performance is pretty much as poor as ever.



Now for what I get when the drives are seperate volumes and loaded into Win 7 as JBOD and software stripe:



So obviously there has to be something wrong with the LSI hardware RAID as especially those 4K numbers with 64 queue depth are just horrible (less speed than a single drive).

Thoughts on what can be wrong with the LSI hardware RAID? Both tests are obviously with the same LSI chip, drives and firmware, so not sure why the large difference.

[mobilenvidia]

Why is AIDA64 showing the SAS2208, is this the LSI9265 ? or is AIDA not recognise the new SAS2308 ?

If you want huge numbers, what are your results for ATTO ?
4x drives for me (SAS2008) show 1.7TB/s read, 2TB/s write, you should approach 4TB/s
OEMs base their speeds on (ATTO scores)
All it is good for is the very maximum speeds, that will never be achieved in real life.

[Callsign_Vega]

The SAS2308 is a faster clocked SAS2008, the SAS2308 supports PCIe Gen3, both are zero cache controllers.
In IT mode (passthrough) these are the best for ZFS and other OS's that do their own RAID.
In IR mode these controllers are as good as it gets with simple RAID (0,1 and 10)

Zero cache = you don't get write back, read ahead, Cached IO etc etc.
The only option you have is turning the drive cache on/off, with SSD's turn this off for better results.

Now the Extreme11 board has 2x PEX8747 PCIe bridges 16 lanes in and 32 lanes out each, these 64 lanes head of to the 4x PCIe 16x slots
Now the LGA2011 CPU has 40 PCIe Gen3 lanes, 32 go to the 2x PEX8747's, leaving 8 Gen3 lanes spare.

Now I can't be certain if all the 8 Gen3 lanes go to the SAS2308 or if other devices share lanes with it.
There are also 8x Gen2 lanes from the PCH, not sure how these are used either.
I do hope that the SAS2308 gets the 8x Gen3 lanes, even 8x Gen2 would be OK.
But if it only get 4x lanes then you will bottle neck as above.
There is a Datasheet on the Extreme11 board layout somewhere which has it drawn, can no longer find it.

LSI has the new P14 Firmware for the LSI9207 SAS2308 controller.
You should be able to update to this, as the onboard BIOS look the same as LSI's.
You will need IR firmware to have RAID.
You can upgrade to this via MSM in windows, load the Firmware then load the BIOS

Turn Diskcache OFF! in MSM, don't touch the Device manager Drive settings, default is king.
REBOOT for setting to take.
To help diagnose run a single drive test first, this is your base speed.
Setup a RAID0 array with 2 drives, REBOOT, get performance figures for this.
Keep adding a drive and do more tests.
At a certain point you may bottleneck, theoretically each drive you add should add the single drive performance.
Theoretically you should be able to get 8x ~500MB/s = 4TB/s

If the single drive speed is already slow then something is afoot, adding more drives ramp as slow.

When you get the speed sorted you'll be able to shift this to the eXtreme section

Would love to do a review on this powerhouse.

I'm positive there are no bottlenecks with the way the LSI chip is connected via the PCI-E 3.0 8x lane to the CPU. 16x goes to 1st PLX, 16x to second, leftover 8x goes straight to LSI. Normal X79 connects to CPU via DMI 2.0, not PCI-E. I have flashed my 2308 to P14 and it now reads 9207-8i instead of "eval board". This produced no greater speeds. I also have the latest BIOS.

I turned diskcache off in MegaRAID and rebooted. Zero speed change.

it will not go as high in RAID, simple as that. if you were to configure the device as 8 individual volumes, then assign a worker to each, you will not be able to saturate that controller. as said above, this is intended for OS's that do their own RAID, and the windows RAID is horrible. that is a limitation of windows, not the hardware.
I can tell you that this is the setup that they were using with the video, 8 different volumes.
I have been able to achieve similar results, and my review is in the queue at TT to be out soon. IR just doesn't have that much sauce in it, but it is more for mirroring than anything.
bear in mind, this is an HBA, and a HBA chipset. you are expecting more than it will give you in RAID. Many HBAs dont even have integrated RAID.

people on other forums telling you that it is a bargain basement raid chip..... well its an HBA, what do you expect?
saying the chip is no good is inane. It is the highest performing chip on the market in its class, period. in passthrough its IOPS performance surpasses that of every HBA and/or RAID controller out there. how is that "no good?"

From reading at the [H] forum thread as well a few things....

You need to delete the firmware from the 2308 before you can switch to IT firmware. that is why it says you " Cannot Flash IT Firmware over IR Firmware!"

second..a few others have told you the key as well, configure as separate volumes, but yoiu have not tried. You have to understand that IR RAID basically 'boosts' the RAID into the OS, where some of the work is transfered over to the OS via the HBA driver. IE....it will suck just as bad as windows RAID.
if you want the numbers do as the review that you have read states, and many have suggested as well....separate volumes!

I really appreciate you guys letting me pick your brain as I just got into this RAID stuff. I guess I was a bit naive thinking with this board/chip I could toss 8x SSD's in and get really high speeds minus a bit of overhead. I guess I am used to the Intel PCH RAID that is simple to set up two drives and is fast.

I set up each volume individually and assigned each volume a worker with 32 QD just as you said. But the test just sat there for an hour saying "Preparing drives" as it painfully slow filled up all of the test volumes 1 at a time. If I let it continue the test would have taken three hours. Does that sound normal or is there something set up wrong in my IOMeter?

I guess I also don't see the point of testing 8 individual volumes unless you are looking for a bandwidth limitation or something. Running 8 individual volumes defeats the purpose of what I am trying to do. (get the most speed from a usable configuration). Preferably a hardware RAID setup that is bootable with Windows on it. If the hardware RAID sucks on the 2308, I'd have to buy two more Vertex 4's to put on the 6GB Intel ports to run Windows and them put my App's/games on the LSI software raid. Kinda not what I was looking for.

Can I ask if as you say the IR RAID firmware doesn't work that great, why didn't they put the IT firmware on it? Mobile said above you that the 2308 is as good as it gets with simple raid volumes, so that's why I'm a bit confused. Some say it's good some say it not.

Do you think I should flash to IT? Will that perform better ina JBOD Win 7 stripe volume? If Win 7 stripe stinks, is there software that I could buy that would run the software RAID in Win 7 a lot better? The hardware RAID seems to fall on it's face when queue depth increases, as shown on the 4k-64Thrd tests versus Win 7 software RAID.

I can try the seperate volume test again, I just thought it very odd that a HDD test program would fill up all the drives and take hours to run a test.

Why is AIDA64 showing the SAS2208, is this the LSI9265 ? or is AIDA not recognise the new SAS2308 ?

If you want huge numbers, what are your results for ATTO ?
4x drives for me (SAS2008) show 1.7TB/s read, 2TB/s write, you should approach 4TB/s
OEMs base their speeds on (ATTO scores)
All it is good for is the very maximum speeds, that will never be achieved in real life.

Ya, just might be reading the metadata wrong or can not identify the 2308 yet. These are my ATTO scores (default settings), let me know how they look (they are higher than AS SSD as you explained):

8x JBOD Win 7 stripe:




8x LSI RAID:

[Computurd]

I set up each volume individually and assigned each volume a worker with 32 QD just as you said. But the test just sat there for an hour saying "Preparing drives" as it painfully slow filled up all of the test volumes 1 at a time. If I let it continue the test would have taken three hours. Does that sound normal or is there something set up wrong in my IOMeter?

sorry, you should configure it to create a 10GB test file on each drive. it was just filling drives to 100% if you do not specify. In the "Maximum Disk Size" put this value: 20971520 for each of the workers. I configure one worker, then hit the "start a Duplicate of this worker on this manager" button up top to spawn a few more identical ones (8).

a better software raid program is needed, for sure, i am just unaware of which would work better.

[Callsign_Vega]

Compu, Ok cool I will try that. As for the IT firmware, do you think it would gaurantee faster software RAID results? If not I think I'll pass. I tried using sas2flash to erase the firmware but it said I could not do it on this platform. Did some research and apparently it will only work in DOS boot disk. Then I found this:

http://kb.lsi.com/KnowledgebaseArticle16266.aspx

"5.DO NOT REBOOT. If you do reboot, or if you attempt to flash the firmware and/or BIOS image and it does not flash correctly, you will have to RMA the controller."

Sounds a bit risky! Especially for unknown results. What do you think is it worth it to try or not? Don't really feel like RMA'ing the motherboad lol.

BTW what do you think of the ATTO scores above. Good/bad?
Oh and I forgot to ask in IOmeter should I set that 10GB test file to 100% seq only?

[Computurd]

i would not try it. tbh it isnt worth it. and it makes RAID not work, anyways. it is just a vanilla firmware. the ATTOs are pretty good, for RAID that is about the highest you are going to get.
With Iometer you can test with 128K sequential at QD32 each disk, and then also 4k Random @QD32.

I think the HBA in raid is going to fall short of your expectations, and i know that sucks. Im just trying to guide you in how to make some eyecandy screen shots

for real world use that RAID in IR mode will be stanky for a boot volume. I think that is the jist of your use, is kinda like an ultimate gaming machine? this will be so much more than you will ever need, trust me ive built a few of the same!

Ive butted my head against the same wall as you with HBAs for sure. and with some RAID cards!

[mobilenvidia]

With your ATTO score you can see where Asrock got the 3.8GB/s from in their blurb.
That score is what I would be expecting with 8x SSD
But meaningless as you'll never get that in real life.

So as Compu has been saying and now that I've seen it, the SAS2308/2008 are just not meant for IR mode with SSD's
The controller CPU simply can't keep up with the data coming in.

You may as well run the SAS2308 in IT mode and run the drives with SW striping
Or run in IR mode just for RAID1

Hmmm, lots learned in this thread, LSI has some work to do with their HBA's once SAS3/SATA4 comes out, their current HBA's are going to struggle

[Callsign_Vega]

Haha I really apprecate your guys help. I know this is complete ridiculous overkill for a gaming machine, but I like to dabble in different stuff for extreme speed and this motherboard has allowed me to go a bit extreme on the storage side. All in good fun.

Here are the IOMeter results individual volumes QD32:

Seq read:


Seq write:


4k read:


4k write:




Now for JBOD Win 7 stripe:

Seq read:


Seq write:


4k read:


4k write:




And lastly, LSI harware raid:

Seq read:


Seq write:


4k read:


4k write:




So just as suspected, fastest goes from single volume's all tested at once, to Win 7 stripe, to hardware RAID. With a huge drop-off on the hardware raid 4k numbers at higher QD. Not sure if it's just getting overwhelmed or what.


So for the "ultimate gaming machine" type setup, would you run 8x V4's in bootable hardware raid with Win 7 on it, or should I get say two more Vertex 4's to run intel RAID on the 6GB port's for Windows 7 and install all the app's/program on the 8x V4 Win 7 stripe volume?

This is the speed I got with two V4's on the Intel RAID that Win 7 would be installed on (write cache on):




I take it you guys are leaning towards the 2x X79 boot-OS / 8x Win 7 stripe setup as the LSI hardware RAID doesn't appear to be that good...

[Callsign_Vega]

OK, so I broke down and flashed the LSI to IT. Got a nice little speed bump! Win 7 stripe almost as fast now as individual volumes (all tests once again 32QD):


Seq read:



Seq write:



4k random read:



4k random write:



AS-SSD:



Blows the hardware raid out of the water.

[mobilenvidia]

A while back I did an article on the LSI9240 (SAS2008) and its very lackluster RAID5 performance due to lack of Cache to do its thing.
I did recommend the LSI9240 crossflashed to a LSI9211 (SAS2008 HBA) for SSD use.
At the time it my 4x Solid3 performed really well on it.

Now you have just shown the best way to use the SAS2308 HBA is via IT mode and Win 7 or other OS striping.
The RAID capabilities of the SAS2308 should be confined to Spindled drives only.

Thanks for the ride

[Snuddi]

The RAID capabilities of the SAS2308 should be confined to Spindled drives only.

Sorry for Hijacking your thred CallsignVega

I am not sure that the SAS2308 is even good for that [Spindled drives only]. As my results with 8x 1TB Spindle drives is horrible with this chip on this motherboard.

Here are my drives before on my Adaptec 6805e raid card in RAID10



And here are the same drives in RAID 10 on the Asrock Extreme11 with the LSI chip....



So with those speed there should not be and overload or anything on this controller.......so I am not sure what is going on.