(SSDs) ICH9 vs Adaptec 5405 - anyone interested?

I've looked around a lot over the past few days and there seems to be very few direct comparisons between onboard raid and external raid with regards to SSDs. At least I was not able to find the answer I was looking for - Is an external raid card really worth it for Raid0 SSDs (ofcourse dealing with low # of drives)?

I have two OCZ Core V2 60GB SSDs sitting in Raid0 on a very clean OS on ICH9 right now. I just ordered an Adaptec 5405 card (256MB cache). I can run all sorts of benchies on the ICH9 and then repeat them on the 5405 after I reinstall to see what kind of benefits that $450 investment will bring when looking at random writes, reads + writes, etc. I am also looking to compare some real world stats like windows boot time and loading times for games.

If some people are interested, I can do all that and stick the results in some sort of table for you all to see. If you want to see this then start shouting out benchmarks and I'll do them. If you are going to say IOMeter then specify all the settings that you want to see. That goes for any benchie that is not very straight forward.

I would be very interested in these results! Just what we need.
I hope the testing won't stop at only synthetic benchmarking. :)

Doing some large app. installs on these setups while they are running OS and/or some downloading (as in writes to disk during the installs) would also be nice.
As well as som large file and lots of small (jpg sized) copying around the volume would also be indicative of R/W performance and general usability IMO.
:D
Making a table using all the compiled results for real life and synthetic test would also be great. (easy to save the table for future reference) ;)

That's all I can think of now... :)

I'm definitely looking forward to seeing some results.

With Core SSDs so cheap... I'm really considering them for storage of games such.

I'll probably keep my 4xRaptors in RAID0 for my OS if the writes are still terrible.

I'm curious how the perc 5i performs with them as well :s combined with the 30GB drives it could make for some breathtaking performance "on a budget" which considering we are talking about SSD's is fairly awesome :D

i would be interested as well i seeing the results because these mlc drives are getting cheaper and have decent performance as well

Id like to see some "REAL WORLD" benchmarks. Grab a stop watch and time how long it takes apps to load and files to transfer.

measured some game loading times (just doom3, fallout3, and fear) and windows boot/shutdown speeds. Also hdtune, hdtach, iometer, pcmark05, ATTO, file copying (stopwatch, 10k small files, 1.5gb overall), game installing(stopwatch, fear and doom3). Got my 5405 and realized I need the cable for it which is not even available anywhere in my worthless country so I have to get it from US which might take a few days(DOH!)

Quote:

---

Originally Posted by One_Hertz

measured some game loading times (just doom3, fallout3, and fear) and windows boot/shutdown speeds. Also hdtune, hdtach, iometer, pcmark05, ATTO, file copying (stopwatch, 10k small files, 1.5gb overall), game installing(stopwatch, fear and doom3). Got my 5405 and realized I need the cable for it which is not even available anywhere in my worthless country so I have to get it from US which might take a few days(DOH!)

---

Plenty of cables in canada, brand doesn't matter.

http://www.frontierpc.com/ProductDet...603&name=3ware

http://www.ncix.com/products/index.p...ufacture=3Ware

Quote:

---

Originally Posted by dengyong

Plenty of cables in canada, brand doesn't matter.

http://www.frontierpc.com/ProductDet...603&name=3ware

http://www.ncix.com/products/index.p...ufacture=3Ware

---

oh god damn it. Frontier PC doesnt have any in stock, but NCIX does, which is just about the only place I didnt check as those guys are normally so overpriced that I do not even bother to use them. Well, too late now. I should have my cable tomorrow or the day after.

I doubt there will be a big difference. Maybe lower CPU usage if anything. 2 drives doesn't saturate the bus enough to bottleneck anything. External RAID controllers are only worth it if you are doing parity calculations or have enough fast drives to saturate the bus. Your money would have been better spend on more SSD's.

Quote:

---

Originally Posted by killermiller

I doubt there will be a big difference. Maybe lower CPU usage if anything. 2 drives doesn't saturate the bus enough to bottleneck anything. External RAID controllers are only worth it if you are doing parity calculations or have enough fast drives to saturate the bus. Your money would have been better spend on more SSD's.

---

This logic doesnt apply to MLC SSDs as that 256mb cache is supposed to do wonders for the extremely low random writes. And you wont saturate the bus of the onboard ICH9R controller. It has the same bandwidth available to it as the external raid cards. I am definitely getting two more SSDs or if these OCZs prove to completely and utterly fail (like they do now) I'll end up going with 4x of the intel ones after their price drop (and stick the OCZs into GF's computer), which is why I bought a relatively expensive raid card capable of doing 800+mb/s.

Quote:

---

Originally Posted by killermiller

I doubt there will be a big difference. Maybe lower CPU usage if anything. 2 drives doesn't saturate the bus enough to bottleneck anything. External RAID controllers are only worth it if you are doing parity calculations or have enough fast drives to saturate the bus. Your money would have been better spend on more SSD's.

---

I've done plenty of tests and I have to disagree.

You're right for spinning disks, but not SSDs.

Quote:

---

Originally Posted by One_Hertz

oh god damn it. Frontier PC doesnt have any in stock, but NCIX does, which is just about the only place I didnt check as those guys are normally so overpriced that I do not even bother to use them. Well, too late now. I should have my cable tomorrow or the day after.

---

Sorry about the frontier pc link, the place I got it from said they had 13 in stock as well as 2 other lengths.

got my cable. Will get home from work in about 6h and reinstall widnows and run all the benchies. If you want anything else compared other than what I listed already, now is the time to speak up.

Ah, I hope the results will be up tonight.

I'm curious to see how ICH9R performs with an SSD RAID.

sorry guys I ran into a BUNCH of problems, I need to go sleep now, but I got a few benchmarks in...

Rough format for now:

ICH9 4k random writes - 7.49 iops, 3569.2ms max responce
ADAPTEC 4k random writes - 24.62 iops, max responce doesnt work right for some reason
ICH9 4k random reads - 4199.44 IOPS, 226.0ms max responce
ADAPTEC 4k random reads - 3077.85 IOPS, 224.7ms max responce
ICH9 iometer seq writes - 172.8MB/S avg, 182.3MB/s max, 225.9ms max responce
ADAPTEC iometer seq writes - 177.4MB/S avg, 180.7MB/s max, 206.5ms max responce
ICH9 iometer seq reads - 268.95MB/S avg, 275.34MB/s max, 228.5ms max responce
ADAPTEC iometer seq reads - 271.44MB/S avg, 273.44MB/s max, 62.6ms max responce
ICH9 1.07GB, 5024 files - 42.8 seconds for copy/paste
ADAPTEC 1.07GB, 5024 files - 8.5 seconds for copy/paste
ICH9 62.2MB, 15,935 files - 32.5 seconds for copy/paste
ADAPTEC 62.2MB, 15,935 files - 7.7 seconds for copy/paste
ICH9 silent hill 4 install - 102.2s
ADAPTEC silent hill 4 install - 55.6s

Look at the real world copy/paste rape! Game install speed got basically doubled too! I am doing everything 10-15 times with reboots and averaging the results so it is no mistake. I have no idea why I lost 25% read IOPS going to the 5405... will investigate later. I have a whole bunch more benchies to do, which I will do tomorrow and put everything into a nice table.

What I would like to see is consistency with read and write speeds. Drop HDTune, for example, on the arrays.

Other than that--those numbers look pretty good. Did you measure CPU usage during each benchmark?

Nice effort, keep the numbers coming. I Like the copy and paste times. Some game and app load times would be sweet as well.

i read a post some ware on the forum some one said if you raid 2 cheap ssd,s such as the core series on a raid card that has 128mb of cache or better you can eliminate the drives from stuttering....true?....i was looking at raid cards at new egg and i honestly cant tell which ones are good or not....how much do you have to spend on a descent card

I've read some good things about this card. About the best price to performance ratio.

Adaptec 2405 Quick price check.

Quote:

---

Originally Posted by VOID WARRANTY

i read a post some ware on the forum some one said if you raid 2 cheap ssd,s such as the core series on a raid card that has 128mb of cache or better you can eliminate the drives from stuttering....true?....i was looking at raid cards at new egg and i honestly cant tell which ones are good or not....how much do you have to spend on a descent card

---

Yes, ALL stuttering is gone. I can actually use my machine while doing write heavy things like game installs. On the ICH9 it would just freeze the whole system and even without doing any writing it would stutter for no reason as well... none of that on the raid card. Will post a complete comparison table tonight.

Looks like the adaptec controller does not suffer from the latency bug of the mlc+jmicron.
Good :)

Nice to things rolling good!
Looking forward to seeing more!
:)

done. This took an absurd amount of time. Basically, if you are going to get OCZ Core series (and probably any MLC SSD but Intel) you need a proper RAID card otherwise you are wasting your money.

Excellent set of tests!
Thank you, thank you for all the time consuming work!

Copy paste results are INCREDIBLE. An on top of that while using a raid card you can actually do something else while copy/paste/installing is going on in the background.
As far as I'm concerned this story is done.
Bring on the next gen. MLC drives and away with rehashing the same ones as new products...

:)

Thanks for sharing the testing results!

Quote:

---

I am doing everything 10-15 times with reboots

---

Wow. You are very dedicated & thorough!

PS. If you have the time, I would like to see some results with the array filled (e.g. 80%-90% full of random files,etc...) to see the effects. I've seen some results where the performance hit is massive, I think many NAND SSD's exhibit this behaviour so far. Maybe some drives have allocated extra space for the swapping/garbage collection when OS visible capacity is near full ?

Cheers!

Quote:

---

Originally Posted by NeedMoMegaHurtZ

Thanks for sharing the testing results!



Wow. You are very dedicated & thorough!

PS. If you have the time, I would like to see some results with the array filled (e.g. 80%-90% full of random files,etc...) to see the effects. I've seen some results where the performance hit is massive, I think many NAND SSD's exhibit this behaviour so far. Maybe some drives have allocated extra space for the swapping/garbage collection when OS visible capacity is near full ?

Cheers!

---

I am aware of that so all tests were done with the array filled up to the max with the iometer test file.

Quote:

---

Originally Posted by One_Hertz

I am aware of that so all tests were done with the array filled up to the max with the iometer test file.

---

Good to know!

So was there any performance increase then when the storage was near-empty? at least on the ICH9R?

Obviously the 5405 dramatically improves performance, but I'm wondering how this affects the analysis at different filled capacity points? Does the controller/cache compensate for the simultaneous mixed read/write poor performance when at full capacity?

Nice results & setup eitherway,

Cheers

Quote:

---

Originally Posted by NeedMoMegaHurtZ

Good to know!

So was there any performance increase then when the storage was near-empty? at least on the ICH9R?

Obviously the 5405 dramatically improves performance, but I'm wondering how this affects the analysis at different filled capacity points? Does the controller/cache compensate for the simultaneous mixed read/write poor performance when at full capacity?

Nice results & setup eitherway,

Cheers

---

To properly do that, you need to send the security erase command to the drives. Once anything has been written, even if you erase it, the performance will stay the same. I have the ability to send secure erase, but of course when I had the ICH9 that was not possible unless I was going to do an extra reinstall of everything, which I was too lazy to do.

"to lazy" would not be the word I would use in any context with anyone who has posted what you just have.
;)

can you also run hdtach or hdtunereally quick?. I would like to see how if there is any difference

Quote:

---

Originally Posted by WRC

can you also run hdtach or hdtunereally quick?. I would like to see how if there is any difference

---

I ran hdtach and there wasn't any difference in read speeds (both setups were 288mb/s, but the 5405 adaptec card had 0.3ms response while the ICH9 had 0.2, which explains the read IOPS drop).

I ran through the thread again but didn't notice - did you make sure the Intel setup had Write caching turned on? It seems Adaptec helped exactly with tests that do writing, and obviously this would make a difference.

Great tests, thanks.
I'll wait the ones with the intel x-25 now :)))

Quote:

---

Originally Posted by alfaunits

I ran through the thread again but didn't notice - did you make sure the Intel setup had Write caching turned on? It seems Adaptec helped exactly with tests that do writing, and obviously this would make a difference.

---

Of course it was on :)

interested in knowing comparison with Intel...........
when do you think you'll have new disks?

300 re-boots, 1min, 300min, 5 hours of re-boots not included time for application runs.

That is xtreme :D

Thx.

Very nice results - have you flashed the 5405 to the latest firmware? It specifically improves read performance ;)

What's the bottleneck for on-board controller?
The cache or the throughput on the pci-e raid controller ?

ICH10R is capped around 650MB/s throughput which is rarely a problem. It doesn't have dedicated cache like a hw raidcon does, meaning you'll still get the stuttering issue with Jmicron & MLC based drives if you run them on the ICH.

Quote:

---

Originally Posted by jcool

ICH10R is capped around 650MB/s throughput which is rarely a problem. It doesn't have dedicated cache like a hw raidcon does, meaning you'll still get the stuttering issue with Jmicron & MLC based drives if you run them on the ICH.

---

Does anyone know how much memory you'd need on the raidcard per drive? Say ____ for 2 drives, ___ for 4,6,8, etc? I know its probably not linear.

The Adaptec 5405 has 256mb ddr2 onboard. Would 128 suffice for 2 or 4 drives? That's the kind of question I am asking.

Quote:

---

Originally Posted by jcool

Very nice results - have you flashed the 5405 to the latest firmware? It specifically improves read performance ;)

---

Nope, didn't have the December one at the time of those benchies :(. For some reason its not letting me do the flash in windows at this time (the button is grayed out)... I'll get it done before I put the Intels in.

Flash in Windows.. are you crazy :D
I always flashed mine in DOS from a bootable USB key, works a treat.

@Synthetickiller: Hard to say really, more is always better ofc. But even 128MB will suffice to run a raided pair of MLC SSDs without encountering the stuttering issues and slowdowns.

Quote:

---

Originally Posted by jcool

Flash in Windows.. are you crazy :D
I always flashed mine in DOS from a bootable USB key, works a treat.

@Synthetickiller: Hard to say really, more is always better ofc. But even 128MB will suffice to run a raided pair of MLC SSDs without encountering the stuttering issues and slowdowns.

---

Thanks. Unless new controllers come out within the next year, I'll probably get a couple of 30 gig drives and a good raid card. :up:

Quote:

---

Originally Posted by jcool

Flash in Windows.. are you crazy :D
I always flashed mine in DOS from a bootable USB key, works a treat.

@Synthetickiller: Hard to say really, more is always better ofc. But even 128MB will suffice to run a raided pair of MLC SSDs without encountering the stuttering issues and slowdowns.

---

Isnt it possible to test >256MB files in atto? Say one 128MB and one 384Mb?

Quote:

---

Originally Posted by One_Hertz

Nope, didn't have the December one at the time of those benchies :(. For some reason its not letting me do the flash in windows at this time (the button is grayed out)... I'll get it done before I put the Intels in.

---

you probably don't have admin rights set-up on your card,
that's usually why you're not able to flash with ASM..........