My results with eight 30 GB vertex, 1 MB stripe
http://i517.photobucket.com/albums/u...EVERPERIOD.png
Printable View
My results with eight 30 GB vertex, 1 MB stripe
http://i517.photobucket.com/albums/u...EVERPERIOD.png
Nice!
More benches please!
Nice Work!
Could you please run AS SSD Benchmark? Just curious what that would show.
Do you have COD4 by any chance? Any games at all?
oh i have many many games, but no fan of the cod's, dont like FPS i prefer RTS!
AS SSD sucks with raid arrays
How about PC Mark, Crystaldiskmark-1000MB, FC-Test
Any problems/quirks with the controller or getting it set up?
Thanks for posting
what is the latency and why is the minimum linear read speed so low?
TBH i dont know why on that run it has such a low one on the minimum..
http://i517.photobucket.com/albums/u...e1/LATENCY.png
You can PRETTY PRETTY please run the fulltest.icf found here http://www.ocztechnologyforum.com/fo...32&postcount=1
w/ a '# of outstanding I/Os @ 64'
I can't seem to find anyone who is willing to run it.
Anyone know if this controller with the current FW/Drv supports the ATA trim command on W7 at least to a single SSD drive (wich supports it of course ;)) ?
best regards TE
I think the answer to this currently is no.Quote:
Originally Posted by dragmore
However i'd expect a new driver / firmware release around W7 launch day that will.
i hope all of you get/buy/acquire this lsi sas thingy..
1231 <- :rofl: :rofl: :rofl:
:D
You have yet to prove that napalm, i do not see anything that you have done to convince me of this yet...other than a test on a fast proccesor running XP...cute video and all but explain?\
Wheres the beef man? Show me some bootup i/o meter runs....
*coughtrollcough*Quote:
Originally Posted by NapalmV5
I agree the hardware has some firmware bugs, but seriously Napalm, what do you hate about it so much? The fact that it's based on a processor more advanced than the Intel IOP 34x line? I think I can fairly say that, given that the 8xxx series was on par and this is the next generation.
I'm sure Intels next generation will also be a beast - but that's what this is competing against. Not last gen's hardware.
INTELS next generation raid controller is based off of the same ROC as this LSI card is..the very same chip!
I couldn't believe it until I looked it up for myself. Intel's next gen RAID controller is based on a PowerPC chip? Lawl.Quote:
Originally Posted by Computurd
But yeah, if even Intel is abandoning ship for this thing... wow. I personally still wouldn't get one yet because I prefer to wait a firmware revision or two, but if the last 5+ years have been any indication an investment now might well keep its value for a long time. Of course, storage is changing quite a bit right now, so I could very well eat crow when saying that, but only time will tell.
It's worse than you know.Quote:
Originally Posted by Computurd
Same chip.
Same card.
Same ram.
Same driver.
Same Firmware (very nearly)
Same windows tools.
Check out the downloads section + pics
Same card as 9260-8i in all but name.
Same price too?
+ one for PCMark05 HDD General Use and Virus Scan test.
Great results so far and for $300 (4i) it cannot be beaten. Thanks for sharing COmpu.
its SAS controller.. its not native sata.. it dont matter if its next generation.. its still soft based+increased access/latency
i dont hate it.. you guys just dont get it.. i thought i was hard in the head but some you just take the cake!!
@ real apps the 1231 still faster.. prove me otherwise.. with your 9xxx you guys should loadup stuff much faster than my 1231 going by your benches and hypes
Intel has no next generation IOP. Intel has sold PXA to Marvell. Which is why the Intel card is exactly the same as the LSI.Quote:
Originally Posted by Serra
OH and guess what the next generation of arecas is going to be sporting fellas? The very same marvell chip!
Proof is in the pudding, or video rather, pay attention when he talks about who the ROC is being made by....
http://www.break.com/usercontent/200...09-747222.html
IF that doesnt prove the legitimacy of this controller as the next gen, i dont know what will! Even areca is using the same ROC!
The areca 1800 series will be sporting the same chip~!
Are you sure the LSI is using the Marvell chip? I am pretty sure the LSI uses their own chip based off a PowerPC design. Marvell's XScale uses a ARM processor.Quote:
Originally Posted by Computurd
Quote:
These SAS controllers are based on the LSISAS2108 RAID On-a-Chip
(ROC) device. This device is compliant with the Fusion-MPT™
architecture and provides a PCI Express x8 interface.
well it does use an ARM based ROC but i could be wrong about it being the same as the marvell.....:shrug:Quote:
Based on industry
standard ARM® processor technology, the LSI Fusion-MPT technology
supports Ultra320 SCSI, Fibre Channel, and Serial Attached SCSI (SAS),
Man, again - put up or shut up. Opening a bunch of YOUR applications doesn't prove **** about the 1231. How the **** are we meant to compete with that? Open a few hundred apps of our own, then enjoy the fact that it's not a fair comparison at all because the apps wouldn't be the same and the fact that opening applications does just plain place a load on the CPU? Add into that that you have a high OC on your C2Q, and others with (eg.) and i7 would be faster and therefore also unfair? Tell you what, at the very least, re-run the test with your proc clocked down to stock frequency and then re-run the test. If your results are any different for the worse, you lose all right to speak, OK?Quote:
Originally Posted by NapalmV5
Seriously man, tests like IOmeter are used in industry to test production servers by many Fortune 500 companies. Yes, they're "synthetic", but they also actually do what they say they do - namely read and/or write files of various sizes. At the end of the day, that's all real-world tests come down to for storage purposes.
LSI users:
Try pcmark vantage!
http://www.futuremark.com/community/...pcmarkvantage/
I hope you can beat my 24/7 computer with Areca 1680ix + 5 vertex :)
If the LSI can not beat the hdd numbers in my run, there is no hope for LSI :p
I hope LSI will fly high! :D
hey Nizzen i will compare with yours, but run JUST the "HDD SUITE", then we will compare those scores. ONLY HDD.
Yes only the hdd suite :)
LSI users: Where are the Pcmark vantage results? :shrug:
Or PCMark05 HDD Suite.Quote:
Originally Posted by Nizzen
No Vantage FTW :)Quote:
Originally Posted by Jor3lBR
OK, here is pcmark 05 hdd tests:
http://i413.photobucket.com/albums/p...cmark0528k.png
Here is my vantage results....
#9 in the hall of fame
FIRST RUN
http://service.futuremark.com/compare?pcmv=222666
dunno much about this bench, long workday today, i guarantee that i will give you fullresults tommorow its my day off!
http://www.futuremark.com/community/...pcmarkvantage/
#9 baby! Exhilirating!
Very Nice Computrd :clap:Quote:
Originally Posted by Computurd
Could you please post your PcMsark05 also cause I can compare with my ORB link.
I just made that run with the OS on the array, there are some improvements that need to be made according to my sources...LOL...anyways that was a prelim run of vantage, gonna actually set it up with the uber 1 MB stripe, this controller loves that baby! THEN i will get to work...i will post that for you jor3lBR by this time tommorow, promise!
@NIZZEN....it has come to my attention that some of your results are not submittable, those are skewed results, your XP Startup benches show that those are 'borked' (i.e. defective) results. You are aware of this. Why are you posting these unsubmittable results!!??
Quote:
Originally Posted by Computurd
I just ran pcmark 05. I can post a another result later today :) But the Pcmark Vantage result is good anyway :D
ok, thanks! That is a very very nice pcmark vantage score, wow!:eek:\
gonna do my best to challenge it!
1231/4x jmicron slc ssd
http://img355.imageshack.us/img355/4111/1231bc8.jpg
yes this bench is not playing nicely with this controller...i am getting real random results, i need to take it off this nf200 slot and run more benches but that screws my tri-sli...the thing is this benchmark is giving way lower numbers than any other benchmarks, running it in win7 though, that may have something to do with it...
Running windows 7 here too, and it is fast :D
Yes, but there are a lot of optimizations that go into running these benchmarks, and guys who do the hwbots are very very familiar with that stuff, i am not..:mad:
The differences between these controllers are not that big, they are not scaling in these vantage and pcmark05 as they are with other benchmarks. A ten percent difference in i/o meter, etc, is a thirty percent difference in these benchmarks, must be something to do with how they bench....
Quote:
Originally Posted by Computurd
I guess there is an amazing tweak some guys are doing to get these results on HDD General Usage and Virus Scan:Quote:
Originally Posted by NapalmV5
http://service.futuremark.com/compare?pcm05=1987222
There is just no way!
I beat him on every single HDD tests but those two and strangely only those 2 gave him the most points.
Here's mine run (ICH10 + 4 Super Talent 32Gb):
http://service.futuremark.com/compare?pcm05=2078712
^ but of course :rolleyes:
no optimizations/tweaks whatsoever.. raw/pure performance
you guys want optimizations ?? here you go..
cache optimization
http://i32.tinypic.com/2gu011x.jpg
not good enough ??
here you go.. mft optimization
http://i25.tinypic.com/168xw9d.png
can your future controller do that ?? :rolleyes:
i think looking to these types of benchmarks, I.E. Pcmark Vantage and PCMark05, to accurately gauge the performance of storage systems and controllers is a fools errand anyway, they are designed for something else entirely! The differences shown between these systems using proven benches, everest.i/ometer/as ssd etc...just dont scale with the differences shown in these other benchmarks. oh well it is fun playing with them though!
EDIT: Just seen napalms little post there...FUNNY:rofl::rofl:
Just proves that you can make it see whatever you want it too, and that running these was a pointless waste of time! LOL
ooyh well that #9 on vantage is still pretty sweet though!
Quote:
Originally Posted by NapalmV5
Holy Bananas!
I hate HWbot now :D
Napalm can you do that and limit the XP StartUp bench at 220MB/s?
I say the only way to resolve this is by running bootup i/o meter, mr nizzen, post it up!
Here is my run when I first tried out my LSI, original FW, poorly optimized, getting better, but I feel this hardware has a ways to go with updates before it shines/ if it will shine.
http://service.futuremark.com/compare?pcmv=214221
Nice!:up:
Great job!
Napalm has done something no one else has done on this forum (expect low fat), he has demonstrated real life performance. You guys call out Napalm to run benchmarks and when he does you start to query the validity of them? :shakes::rolleyes:
Kudos to Napalm. :up:
+1
Testing real life performance is very hard, especially if critics demand a super high scientific methodology... there is none that people will agree upon. It has to be forced on us by some test site guru or forum "moderator" sort to say.
New LSI storage manager up on their site.
Grab it here
Personally I prefer the layout / functionality of the older version but the new one works fine too...
@Napalm - It's good to see some results we can compare between. Could I bother you though to run an ATTO and/or IOmeter as well?
LSI released new drivers, firmware and MSM yesterday.
YaYY!!Quote:
Originally Posted by Anvil
Download link :)
Ok, so I have the 2.30.03-0775 FW
I'm running on a X8DAH-+F Supermicro board with two Xeon 5560 24GB RAM (which will be 72GB when they send me the Registered RAM) and W7 x64 ultimate.
I'm running my SFF-8087 cable from my backplane on my Supermicro SC216 Chasis to my raid controller. My hard drives are Photofast v5 SSD 512GB hard drives. My raid setup:
RAID0
Read Policy: Adaptive Read Ahead
Write Policy: Always Write Back
IO Policy: Direct IO
Access Policy: RW
Disk Cache Policy: Enabled
Background Initialization: Enabled
on a 1M stripe.
My numbers are
READ:1504mb/s
WRITE:1736mb/s
How can I increase these numbers?
Sequential numbers, add more drives, faster drives, or possibly OC the PCIe bus.
The question is, what are your IOPS numbers, and do you NEED to increase storage performance? Or are you bottlenecked elsewhere?
Have you done a comprehensive search for bottlenecks in different situations?
You are aware even a 10GbE connection will be a bottleneck for (the sequential performance of) your array?
EDIT: you may get better results with a smaller stripe, f.ex. 64KB, 128KB, or 256KB.