What's wrong with your .xls files?
What's wrong with your .xls files?
Couple of questions on MegaRAID SAS 9260-4i kit:
1. What type of cable does it come with? Mini-SAS to SATA?
2. What would be the best price/performance SSDs for it? I'm thinking 30Gb/40Gb. Great 4K random reads being priority, and nice sequential reads are second. Would cacheless drives be the best since it already has 512MB onboard cache?
3. Does it support SATA 6Gb/s?
Cheers
1. Mini-SAS to 4 x SATAQuote:
Originally Posted by F@32
2. Intel X25-V
3. SSD interface is 3Gbps but support 6Gbps
Probably linked somewhere above, but this does not look too pretty for 9260-8i.
http://www.legitreviews.com/images/r...andom-Read.pngQuote:
For controllers I have four options available to me. The two non-SAS controllers were the Areca 1220 and the ICH10R integrated into the chipset of my Gigabyte P55A-UD6. The two SAS controllers were the Adaptec 5805 and LSI 9260-8i. Cached writes were enabled on all controllers along with read-ahead when available.
http://www.legitreviews.com/images/r...ndom-Write.png
ummmm that review is old bro. the 4k potential of this card has been expanded tremendously along with its latency in the latest round of firmwares. i think that was published tow firmwares ago. that is something they have alluded to as well in the conclusion of that reeview. this is relatively virgin silicon. a lot of room left for improvement. the FastPath software that is coming is supposed to triple small file random IOPS in some cases.
here is a screen shot of the 4k random file performance from the last firmware. the improvement. also look at the latency improvements. fantastic! (this is from previous testing i have done on this forum) you can see the 9260 matching ich10r.
http://i517.photobucket.com/albums/u.../forsquish.png
I must be getting stupid or something, but how the heck do you install a firmware on this controller?
I downloaded MegaCLI for Windows 7 x64, extracted it, extracted MegaCliSupport.zip, extracted AF2108_FW_Image.rom, all in the same folder.
I disabled UAC (putting the slider to the lowest stand)
and tried:
I also tried editing cliVmWare.conf, but I have no clue what to fill in...Quote:
C:\Program Files\LSI\MegaCLI>MegaCli64.exe -adpfwflash -f AF2108_FW_Image.rom -a
0
CreateFile failed with error: 5
CreateFile failed with error: 5
CreateFile failed with error: 5
ERROR:Could not detect controller.
Failed to get ControllerId List.
Failed to get CpController object.
Exit Code: 0x01
C:\Program Files\LSI\MegaCLI>
Should uncomment "location" and "user", fill in 127.0.0.1 and root?
Or should I leave those 2 commented out?
Also I have no clue where to set the password of it...
I tried logging in with a lot of passwords in the MegaRAID Storage Manager, but no passwords seems to work.
In the documentation I also can't find anything about a default password...
Can anyone help me out?
god i had a helluva time with the CLI as well, wouldnt touch it with a ten foot pole. you need to use the megaraid storage manager (MSM). there is no password if you didnt set one, just use your computers user account name as user name. leave Password blank. unless you set one of course :) lemme know how you fare...
aha
I am indeed able to login in MSM with my windows user/password
Thanks!
So did you configure cliVmWare.conf for flashing the firmware?
sweet!
Apparently I had to reboot to really deactivate UAC...Quote:
C:\Program Files\LSI\MegaCLI>MegaCli64.exe -adpfwflash -f AF2108_FW_Image.rom -a
0
Adapter 0: LSI MegaRAID SAS 9260-8i
Vendor ID: 0x1000, Device ID: 0x0079
Package version on the controller: 12.0.1-0045
Package version of the image file: 12.3.0-0022
Download Completed.
Flashing image to adapter...
Adapter 0: Flash Completed.
Exit Code: 0x00
C:\Program Files\LSI\MegaCLI>
I also completely filled in that .conf file, not sure if that was required though...
thanks for the hints ;)
now that the driver, firmware and software all is the latest version, I might wanne try to configure my first temporary RAID 5 as a try-out
(my 4 Seagate drives are going back next Saturday because of the clicking DOA, but 3 of them still seem to work enough to play a bit already)
has anyone seen this one? :P im just waiting for the next firmware :P
Again LSI pass ARC
If this works soon I say :clap:Quote:
Originally Posted by Tiltevros
:up:
GC is working if ur drives are compatible with it Intels are not so i have to wait for the next firmware.
The best think with this is that it works on virtual disk management
If they can get TRIM added to that list also, it would be a great sales point on their end :)
I don't know if by means of programming it is difficult or undoable, but, yeah, it would save some time from the present maintenance procedure....
@ computurd. get your ass in here asap! I just got the new fw and drivers today.
read here:
http://www.xtremesystems.org/forums/...d.php?t=251365
lol i just got home...this should enable fastpath! but there is a trick to it...
lolQuote:
Originally Posted by Computurd
okay paul, its been a while since we last did this crap.
is this any good? I just did the 9260 fw rebooted and set up the raid using the lsi webbios and used the auto config. installed win7 and did the basic intel chipset, ati, then rebooted and switched to ahci in asus bios rebooted and disabled index, dfrag and superfetch I didnt touch anything else.
system is:
core i3 540 @ 3.374gb [ 160x21) qpi 3534.7]
maximus III formula 1156 (1607 bios)
5850(8x) / (lsi 9260-4i 8x) split x16 8x8 (2 pcie x16 populated)
2 x 50 gb ocz vertex raid0 / 64kb stripe using lsi 9260-4i sas/raid card [Driver = Windows 7 (64-bit MegaRAID Release 4.4) 4.24.0.64 06-MAY-10 / FW MegaRAID Firmware
(MegaRAID Release 4.4) 12.6.0-0017(APP-2.60.03-0829) 06-MAY-10
4x2gb elpida hyper @ 1600 / 7-7-7-24-72 1t
these are using lsi webbios AUTO raid0 config (ctrl H) ncq enbled
http://i239.photobucket.com/albums/f...ttodefault.jpg
http://i239.photobucket.com/albums/f...unedefault.jpg
---------------------------------------------------
heres the lsi config:
http://i239.photobucket.com/albums/f...licies9260.jpg
http://i239.photobucket.com/albums/f...lt/lsimsm1.jpg
http://i239.photobucket.com/albums/f...lt/lsimsm2.jpg
http://i239.photobucket.com/albums/f...msm3config.jpg
http://i239.photobucket.com/albums/f...licies9260.jpg
----------------------------------------------------------------------------------------
now after going into msm in windows I set the rest like we did on indlinx/barfoot
no option for "adaptive read ahead" so I just used "read ahead" etc.. benched again..
see changes:
http://i239.photobucket.com/albums/f...generalncq.png
http://i239.photobucket.com/albums/f...gs/changes.png
http://i239.photobucket.com/albums/f...tings/atto.png
http://i239.photobucket.com/albums/f...gs/hdtune2.png
---------------------------------------------------------------------
side by side compare after changes. NCQ is still enabled
http://i239.photobucket.com/albums/f...ttodefault.jpg
http://i239.photobucket.com/albums/f...tings/atto.png
http://i239.photobucket.com/albums/f...unedefault.jpg
http://i239.photobucket.com/albums/f...gs/hdtune2.png
It would seem you are caching the entire 256MB ATTO test file, or most of it.
Try using a 2GB test file.
Also, AS SSD numbers would be nice.
@trans am...
try using iometer or crystal bench mark. also everest is a pretty good tool for latency testing. also the winsat disk test within windows is about the best imo.
elevate a cmd prompt to admin status. then when cmd box opens up type in:
winsat disk
that will give ya some good numbers there. most of your numbers up there are cache, as gullars points out. atto is trash anyways dont even bother.
Beg to differ, but if you use it for the right purposes and know what you're looking at. It gives you a comprehensive dataset quickly. Just always keep in mind, ATTO usually gives best-case numbers, they are sequential, and default QD is 4, but you can change it.Quote:
Originally Posted by Computurd
I like to use ATTO at QD 2 (it annoys me QD 1 is not avalible) to chart sequential read/write scaling by block size, and to look for possible problems (drops or inconsistencies) as more quantitivly suppliment to CDM and AS SSD. IOmeter is ofc more accurate and can give you more extensive data sets, but it's much more work gathering and enterpreting the data there.
atto is the most inconsistent benchmark i have ever seen. run it consecutively a few times under different conditions and you will see. set adaptive read ahead on just about any raid controller and watch those numbers go crazy.
i guess i should refine my statement. for hardware raid cards atto is trash.
it is easily confused by things such as read ahead, and gives out very inconsistent numbers. any benchmark that gives out widely unreproducible results makes me question it heavily. seems to have problems with ssd in general, more specifically of course the raid cards. it does have an easy to read format, i agree whole heatedly with that. if only it was more reliable. they need to get better testing tools for storage in general though, this has been a big complaint of mine for ages :)
Quote:
Originally Posted by HiJon89
huh??? and what those founctionality be... besides the error reporting cappability of sas .....
....................
http://www.scsi4me.com/lsi00247-fast...9260-9280.html
Okay lemme get this straight we need to pay $150 bux for the software and it offers huge performance boosts? I think its rad that LSI offers this hotrodding soft to add on its existing raid cards.
Paul do you have the software installed and did you see any improvements? I'm anxious to know because I want to crank these new sandforce drives I just got.
thats the thing bro, not available for about twenty more days. kinda BS we have to pay for this.
there is more to the story i guess.....
received this message today
Quote:
The information I forwarded you was incorrect as it is an out of date document. More information will be disclosed after our May 10th announcement on these new additional features. Keep in touch with your distributor on release dates and pricing information as they interact directly with the sales and marketing team.
$150 bux for something which seems to just increase high queue small random reads/writes. I don't see that actually helping anything for a normal PC.
If it reduces latency (overhead) for larger SSD RAIDs, it could make sense for the people buying controllers like the 9200 series.
From the little info available it SEEMS like it only reduces latency in high queue + small file workloads + more than 4 drive R0, which would not be very useful (for us here on this forum).Quote:
Originally Posted by GullLars
We'll see what numbers computurd puts up, that should tell us if it's worth it even for large arrays. I thank him for being the guinea pig.
Today is the day I'm finally going to build arrays on ich, I want to see if there's a sweet spot for c300s at 3 or 4 drives.
Paul - thanks for taking one for the team!
ok here is the misunderstanding..
Im talking to the guy who was at Cebit and demostrating the Fast Path.
Fast Path is a key that unlocks the lower latency algorithm and set power performance on SSD's.
The demostration was with 5 SSD (x25 E's) just to show that u can pass the 150.000 IO's with 8k file size witch means that 4k file size will be more.. This is nothing to do with large files because LSI never had any problems with them.
Now using more than 8 drives will give more than 150k IOs thats why FastPath is compatible with 9260 4 port.
Just give me and Paul some days to get the keys and we will post everything u want
tilt hit me on pm
@ trans am
First of all update ur firmware. u r saying the best performance setup for what??? for databases?? for gaming? for IOmeter?? for PCmark Vantage? for what?
did it all. I have the latest and greatest of both. latest lsi fw/drivers and latest and greatest ssd drives. I wanna talk to you or paul or steve cuz ive been in the gutter and only been benching clarks and not really up to speed as far as the best all around lsi setup in webbios.
i was using the following in indinlinx raid:
2 x indinlinx 60gb
raid 0
9260-4i
raid0
stripe size 64kn
adapive read ahead
r/w
direct i/o
drive cache enabled
bbu off
ncq disabled.
use server 2008 x64 drivers for (win7 64)
my question is:
say I sent you my drives and you wanted to bench them.
what settings would you use when creating the raid set (latest may drivers and fw)
just want someone with me while we max these out.
fyi I just wiped the vertex2 with killdisk and as soon as you get back in here I will set the raid up and install win 7.
this is what I did the night I got it all installed but I think I can do better.
http://i239.photobucket.com/albums/f...ttodefault.jpg
http://i239.photobucket.com/albums/f...unedefault.jpg
http://i239.photobucket.com/albums/f...ngs/ncqoff.png
http://i239.photobucket.com/albums/f...randomread.png
http://i239.photobucket.com/albums/f...gs/hdtune2.png
----------------
intel onboard raid on 2 x 50 gb vertex2
http://i239.photobucket.com/albums/f...atttointel.png
adapive read ahead <--- Doesnt exist from firmware 2,30 now the new firmware is 2,60
ONLY
Always read ahead and no read ahead options
LSI performs best with Intels compatibility list is on the site LSI.com
ur performance is good as i can see.
okay tilt what about stripes? keep it at 64 or go higher or lower? I have adaptive option in webbios (ctrl-H) so I had same options when I set raid0 in webbios as auto and went into msm always showed read ahead only. but if you create in webbios adaptive is still there. which one? which stripe? lets get this rolling!
16k for intels search for other brands :( u should try all of them.
this is the best thing with this controller. Start looking our threads there are millions setups here with 9260 read them ok?
tilt here is my ss from msm http://i239.photobucket.com/albums/f...ith/9260fw.jpg
im on 16kb right now. Gimme a bench to use
also waht about this guys settings? http://i263.photobucket.com/albums/i...8isettings.jpg
IOmeter
Oh man. can you set up a guide sticky for us to bench iometer? Ive been through this with paul so many times. I think it needs to be on paper. there are so many settings I alsways get lost. just make a sticky and we can all follow it and have something to follow once and for all. I'm gonna be out setting up the raid again and reinstalling. so this is what I'll do:
LSI
raid0 128kb
read ahead'
read/write
io direct
drive cache enabled
always write back
ncq disabled
bbu off
You should try stripe size 32KB or 64KB using the Vertex 2.
No read ahead
Direct IO
Drive Cache Enabled
NCQ ON, why would you wan't to turn it off?
Write Back is OK but you could also try Write Through.
BBU off?? <--- if u have BBU set it to write back with BBU
NCQ disable? <--- there r no difference if u have this enable or disable
read/write?? <--- let this as it is
Im testing LSI 9211 right now im waiting some keys for my 9260 this week so in the weekend i will have the time to play around.
But u should read the threads for 9260. To just set the settings is very easy but to hit the best result is differend to each one.
Im gathering some tips and when i will have the time i will post them here
BTW my booting time now with windows 7 ultimate FULL services is 12,21 seconds LSI 9260 with only 4 intels
I dont have bbu so I disable it. I just set to write back. no read ahead? then what? adaptive?
@trans_am: Honestly you need to try every single setting to see what works best with those vertex.
Hi all,
I just tried creating my first RAID5 on my LSI controller.
I used all default settings atm (64KB Stripe Size, Write through since I dont have a BBU).
I just tried a few quick benchmarks.
Read speed seems normal: 40MB/sec min, 320MB/sec max and 196MB/sec avg in HD Tune
But write speed doesnt seem normal at all: 10MB/sec min, 22MB/sec max and 15MB/sec avg in HD Tune
CrystalDiskMark shows similar results...
Now I know that write speeds in RAID5 arent supposed to be shockingly fast, but I did expect around 100MB/sec for a 500 euro controller (I remember the ancient Areca controllers doing quite well in RAID5 write speeds)
I did a fast initialize and read about it that it writes the first 10MB and last 10MB of each disc with 0 and does the rest in the background.
However I don't see any ongoing operations in MSM...
So is this horrible write speed because of the secretly in the background running initialization? (if yes: where / how can I see if it is done?)
Or is this horrible write speed really the speed it should have? (because of the write through setting and RAID5)
I definitely hope something is still wrong somewhere ;)
Thanks
If i'm not mistaken, you need Write-Back Cache to get OK write speeds in RAID-5, write-through give the horrible write speed you see.
Right before going to work this morning, I quickly tried this out and you are absolutely right!!
With Write back enable I get write speeds of around 300MB/sec...
Now I knew that disabling Write back means less performance, but I thought it was something like 60-70% of the performance, not 10%...
Actually that is no longer a loss in performance, but a total destruction of performance ;)
Well... I guess disabling Write back is not really an option in this case, so I need to "go back to the drawing board" and figure out if I now need to buy this overly expensive BBU anyway :(
It seems Tiltevros statement about "not being able to enable write back without a BBU" was incorrect.Quote:
Originally Posted by Mastakillawas that a one time offer at span.com when you got the controller and battery for only 412 euro?Quote:
Originally Posted by Tiltevros
Cause now I see the LSI 9260-8I for 421,42 euro and the battery for 120,90... :(
I have 2 APC SurgeArrests for 8 outlets each to protect all my electronic devices, so power spikes shouldnt be a problem.
Power drops and complete power loss however can still occur.
In case of power loss I wouldnt mind very much to loose what i was working on. However the chance of loosing all data because of power loss causing a corrupt RAID 5 is unacceptable (that is the reason why I'm investing in this expensive RAID controller, I do not want the chance anymore to loose all data. And I'm tired of putting everything on DVDs :p)
So can anyone confirm that I don't need a BBU to ensure that I will not loose all data?
About the write back cache: Interesting... I didnt know you cant enable it without a BBU... Does it matter much in performance? Or where lies the difference?
Thanks!
I just enabled write back cache without having a BBU and got almost 1000% write speed increase...
But my old question about the BBU still wasn't answered... :(
So I'll ask it again:
Is it possible to loose the complete RAID 5 IF
* write back is enabled
* there is no BBU
* a powerloss occurs or a system / application crash occurs
OR will this only cause me to loose the data in the cache (+ a bit of corresponding data that was already on the disc)?
I can only tell you I have crashed my system hundreds of times while overclocking and never dropped my array. I have no bbu and enable write back. Tilt has recommended write back with and without bbu in many other threads so I'm sure he knows you can write back without bbu I'm not quite sure what happened in the post you're quoting.
The answer to this is: both.Quote:
Originally Posted by Mastakilla
It's really quite simple: If you have data you can not lose buy a bbu and backup off site.
You can lose data at any time for more reasons than I could even mention. Your array can fail at any moment. bbu or no bbu, ups or no ups. Ask large datacenter operators, they will tell you 10k$ raid cards fail all the time.
Last but not least your computer could catch on fire and melt all your 010101010's! Life is crazy!
thanks for the answer...
I guess your crashing experience tells me enough to know I don't need the absurdly expensive BBU ;) (price has even risen to 163 euro minimum now????!!!!)
I know I can still loose all data at all moments because of many reasons and I've looked in to tape streamers for off site backups, but they are waaaaay to expensive for the amount of data I'll have
I now have a RAID 5 to have a more secure data system than the RAID 1 (120GB) + RAID 0 (500GB) that I had before (even without a BBU and write back enabled this should certainly be the case...)
btw: I've read somewhere that if the LSI controller dies, you usually can still replace it with the same controller and keep your array
for those interested, below are some screenshots showing the insane performance difference between Write Back and Write Trough on normal HDs:
Write Through:
http://users.telenet.be/Mastakilla/H...ark_wt_1GB.jpg
vs Write Back:
http://users.telenet.be/Mastakilla/H...ark_wb_1GB.jpg
Write Through:
http://users.telenet.be/Mastakilla/H...e_write_wt.jpg
vs Write Back:
http://users.telenet.be/Mastakilla/H...e_write_wb.jpg
Write Through:
http://users.telenet.be/Mastakilla/H...ne_read_wt.jpg
vs Write Back:
http://users.telenet.be/Mastakilla/H...ne_read_wb.jpg
and also a screenshot of CrystalDiskMark set to 4GB with Write Back enabled (to minimize the cache advantage)
http://users.telenet.be/Mastakilla/H...ark_wb_4GB.jpg
as you can see, not only the write speeds increase by almost 1000%, but also both read and write speed stabilize a lot
ive never had a crash on win7 with 9260 in raid0 with wb enabled without a bbu. battery is totally unnecessary. then again im not using raid 5 just 2 sandforce ssd in raid0
I ordered a fastpath key and it comes tomorrow. I cant wait to post results. before and after.
I cant wait to see your results either TA!
The performance difference is because in write back mode, the controller indicates that the write operation is complete after the data has been transferred to cache. (So you're really measuring how fast you can write to cache, not to the drives).Quote:
Originally Posted by Mastakilla
In write through mode, the operation isn't complete until the data has been written onto the drive(s).
This, of course, means that you should have a battery backup unit for the controller if you use write back mode and care about the integrity of your data.
This also means that the perf increase is "temporary" -- it only lasts until the write cache is full.
yup, and there is also a cache flushing interval that will clear your cache every four seconds by default. i set mine to 2 seconds to kinda hedge my bets as i roll without BBU. also adds about 1k points in vantage for some weird reason.
I'm currently running 8x seagate 320GB drives on a 1231ML. I really would like to get into SSD's but won't until trim support is available for a Raid 5 setup so I plan to upgrade to Savvio 15k.2 drives. Would I be better of getting 12 drives and keeping my 1231 or getting a 9260 and using 8 drives?
http://www.dadepc.com/ARC1231_HD.jpg
http://www.dadepc.com/ARC1231_crystal.jpg
From a maximum sequential transfer rate perspective, I think you'd probably do better with the 9260. The numbers I've seen are that the 1231ML peaks at about 800 MB/s sustained.Quote:
Originally Posted by Tutto
If you're going for random IOPS, then with 6 Mbps SAS drives like the Savvio, given that the 1231ML is optimized for 3 Mbps SATA, I'd still go with the 9260.
Do you need 2.5-inch drives? If not, the Cheetah 15K.7 is faster than the Savvio 15K.2 for sequential access: 204 vs. 160 MB/s (the Savvio is slightly faster for random), and you get about twice the capacity for the same price. Eight 15K.7 drives in R0 should be able to get close to 1600 MB/s on a short-stroked partition.
FWIW, your CDM numbers would be more meaningful if you set the size to be at least twice the size of the cache on your 1231ML board.
I have a 1231 I am waiting to see what the areca 1880 looks like before deciding on a replacement.
I dont know why, but when I copy a 10GB file from my SSD to the RAID5 I get a constant speed of 250-290MB/sec and not a constant speed of 15-19MB/sec (the cache is only 512MB)Quote:
Originally Posted by AceNZ
not that I'm complaining :)
With RAID-5, there's an additional benefit of write back caching, which is that the controller can buffer enough data to write full stripes at a time. The alternative is partial stripes, which require reading-then-writing the parity disk. With a full-stripe write, the parity drive is just written, and not read first. With rotating media, there is a huge performance difference, since you don't have to wait for the drive to make a full rotation before writing.Quote:
Originally Posted by Mastakilla
I suppose you mean write back instead of write through?Quote:
Originally Posted by AceNZ
Hi guys,
I'm following this thread with attention, as I'm the new owner of 9260-4i and 4x X25-M.
Here is the performance shot I came to :
RAID0Quote:
-----------------------------------------------------------------------
CrystalDiskMark 3.0 x64 (C) 2007-2010 hiyohiyo
Crystal Dew World : http://crystalmark.info/
-----------------------------------------------------------------------
* MB/s = 1,000,000 byte/s [SATA/300 = 300,000,000 byte/s]
Sequential Read : 906.028 MB/s
Sequential Write : 347.498 MB/s
Random Read 512KB : 675.881 MB/s
Random Write 512KB : 421.801 MB/s
Random Read 4KB (QD=1) : 21.067 MB/s [ 5143.2 IOPS]
Random Write 4KB (QD=1) : 47.653 MB/s [ 11633.9 IOPS]
Random Read 4KB (QD=32) : 166.830 MB/s [ 40729.9 IOPS]
Random Write 4KB (QD=32) : 119.813 MB/s [ 29251.1 IOPS]
Test : 1000 MB [C: 12.9% (38.1/295.9 GB)] (x5)
Date : 2010/06/15 20:04:34
OS : Windows 7 Ultimate Edition [6.1 Build 7600] (x64)
64k Stripe Size
Read Adaptive
R/W
Direct I/O
Drive Cache Enabled
BBU Off
NCQ Disabled
Write Back
Do you have any tips to improve this score ?
Enable NCQ dude. You should see a boost in 4KB random @ QD 32.Quote:
Originally Posted by TuKo
Also try running AS SSD.
BTW, have you got the newest firmware on both the drives and the controller? And have you got the latest LSI drivers?
If you want more performance, the next step is a FastPath key, it will boost your IOPS and give better accesstimes, even at fairly low QDs, and will allow you to get close to 120K 4KB random read IOPS max from your 4 M's.
I got a 9260 with cables for $400 including shipping. Now I have to decide between the 15k.2 and 15k.7 drives. I'm leaning toward the 15K.7 but have to wait for a check to arrive before I make the purchase.
This is my first time using sas drives and want to make sure these are the correct cables.
http://www.scsi4me.com/lsi-07-00021-...pin-power.html
Yes, sorry; I edited my post.Quote:
Originally Posted by Mastakilla
no problemo :)Quote:
Originally Posted by AceNZ
thanks for the explanation
still a bit weird that the difference is so enormous
the benchmarks of other controllers I found show a 40-60% difference between write through and write back, not a difference of 1000% percent :p
It's a great price, where did you buy your card? The best price I find here in Sweden is $890 for a 9260-8i without shipping.Quote:
Originally Posted by Tutto
I won it on ebay. He had it as a buy-it-now or Best Offer. I offered $400 and he accepted. Could you verify if I am getting the right cables for the Seagate 15k.7 SAS drives.Quote:
Originally Posted by 0xdeadbeef
Those cables should work. However, they require connecting to molex-style power cables. With a modern power supplies that tend to have lots of SAS/SATA-style power cables, I prefer the SFF-8087 to mini-SAS style. For example:Quote:
Originally Posted by Tutto
http://www.newegg.com/Product/Produc...82E16816116100
Also, it's possible to connect the 9260 to both ports on dual-ported SAS drives like the 15K.7; your performance will improve if you do.
FWIW, some of the 9260s come with cables (the "kit" SKU).
I just encountered something weird when setting up my RAID5:
I set up my RAID5 in the WebBios using the default settings (except for forcing to always use Write Back).
At the end of the wizzard WebBios asks me if I want to do a fast initialization, so I choose yes.
Then almost immediately I come to a screen where I can again choose to initialize and also do other things (such as a consistency check).
Since I just choose to fast initialize already, I just exited WebBios and booted into Windows.
When I open Disk Management in Windows, it asks which type of partition I want for the RAID5.
I choose GPT and create on big NTFS partition with a custom Stripe size of 64K (same as configured in WebBios)
Then I went to LSI MSM and tried to do a consistency check.
This gave a warning that the disk wasn't initialized yet??
So I thought: perhaps it is still busy in the background... So I left my PC running for 24hours...
Then I tried again, it still sais the disk isn't initialized??
So I try to initialize and it warns I will loose all data. To make it a bit easier for MSM, I then removed the partition and did an fast initialize.
After the initialize, I rebooted (else Disk Manager was flipping), I again created a GPT partition type with NTFS partition (again 64k stripe).
Then I retried to do a consistency check, but it warns the disk isn't initialized??
So I again remove the partition, do a fast initialize and tried to do a consistency check before created a partition.
This time it started, but immediately found inconsistent parities.
So I stopped and decided to a full initialize (not fast), which is running now...
Can anyone explain me what I'm doing wrong? What is the correct way to do it?
I want to be able to do consistency checks while keeping the partitions of course :)
And I know I can do a consistency check on a "non-initialized" disk, but I still prefer my initialized disk to be recognized as initialized as well ;)
After the Full initialization, I did a successful consistency check (so without creation a partition and it didn't warn for the RAID5 to be uninitialized)
Then I rebooted, created a ntfs partition (as before) and then I started another consistency check and again it didn't warn for the RAID5 to be uninitialized (which is still running now)
So the problem seems solved, but I'm still not sure what the problem / solution now was...
It seems like the fast initialization doesn't do its job very well or something
edit: the consistency check with the NTFS partition has succeeded
Hi again...
Now that the consistency check was successfull, I know the RAID5 is currently in good shape.
But I was wondering how I could keep an eye on my indivual HDs themselves?
I havent found an application that was able to read SMART values while they are in RAID5. Also in LSI MSM I cannot find that option...
Can anyone tell me
1) How to check the SMART values of my HDs?
2) Does the LSI controller perhaps have some build in "replacement" for verifying the HDs health?
Thanks!
edit:
I just found the following very interesting page about how Dells PERC controllers keep an array "healthy":
http://www.dell.com/downloads/global...0212-Habas.pdf
This explains the terms "consistency check", "patrol read" and "SMART" pretty well and understandable.
I suppose that LSI controllers work in the same way...
Also I found in the LSI manual that the WebBIOS has an option to change the SMART polling value (didn't find it back in LSI MSM though?).
This indicates that LSI does use SMART to keep an eye on the HDs health.
Only I still would like to be able to see the actual SMART values in some tool. My HDs are officially not supported by LSI, so I would like to check if the LSI is able to read SMART from my HDs...
So does anyone know how to do this?
Gents, do you know if FW is any good on Intel RS2BL040 clone of 9260-4i: http://www.intel.com/products/server...0-overview.htm
It was going for nice discount compared vs. LSI at newegg. Sold out at the moment.
And first impression is quite dissapointing. Specially because i just had Adaptec 5805 on my hand and it performed good.
I still wanted sata3 controller for my Crucial C300 x4 and so decided to get LSI.
To the problem. Look at sequental reads.
http://i29.tinypic.com/oqg70z.png
New firmware didnt help much
New bios didnt help much
New drivers didnt help much
I even installed fresh windows7 on to lsi hosted raid 0. Didnt help at all.
Tried different stripe sizes. Didnt help.
I honestly dont know what i have to do to get a decent 1000+ MBs reads out of this controller.
:HELP!:
Same drives on Sata2 Adaptec 5805 got 1200MBs burst and around 980MBs sequental reads, which is pretty much the limit of 4x sata2 lines.
Now im expecting 1300+MBs from LSI + 4x C300s and cant get even 600 out of it. :confused:
Any ideas guys/girls?
Create a softraid. (generally not bootable)
(means using the drives in passthrough mode and creating the stripe using windows disk management)
Quote:
Originally Posted by Anvil
Ok i got 1314MBs reads with soft raid now. thats good enought. But can you give explanation why controller raid isnt working? Is it some firmware bug or what is it?
I would really get it to work and run win7 from drives.
what is your motherboard? and does it have NF200 on it?
Its Gigabyte 790FXTA-UD5 Running PII X4 @ 3,75Ghz....
Whats NF200?
The controller raid is working, just not to your expectations.
The 9211 is not a sequential performer, it never was, it's great for SW raid and random IO though.
You should have gotten the 9260, I know it's in a different price range but it's what you want/need to buy.
Probably not the answer you were hoping for.
Have a look at the ATTO benchmark, it tells the story.
The max sequential throuhgput is between 128-256. (128 happens to be 2x the stripe size on the 9211 and will always be max throughput)
The benchmarks you are using (AS SSD, CDM) are using a larger block size when measuring the sequential throughput, possibly 1024KB or more.
If you run HD Tune with a block size of 128 or 256 you should be able to reproduce.
You are damn right that its not what i was hoping for. LSI advertises the controller as sata3 controller and 2GBs speeds. All its raid0 good for right now is 500mbs sequentals.Quote:
Originally Posted by Anvil
Then windows softraid can do 1300+mbs through the controller, with my drives. Why the hell controllers raid sucks balls then?
:shakes:
btw. Windows softraid beats controller raid in every single stat. starting from sequentals and finishing with access times.
Im going to write to LSI support. Hopefully they offer some sort of solution.
Big thanks to you Anvil. Without your knowledge i could not figure it out by myself.
its s HBA (host bus adapter), not a true raid card. so its stripe size is limited to 64, being why you are reaching limitations with the speed of course. however, i reached speeds of 990 mb/s with 8 vertexes so there is something going on here with your board imo. would be hard to figure though, i d nt know shat about the amd chipsets. there are other things you can try to maximize throughput. does your board have a maximum pci-s packet setting?
actually my 990 was at randoms, the max sequential was 1024...
http://i517.photobucket.com/albums/u...bootsarray.png
http://i517.photobucket.com/albums/u...tssarray-1.png
so dont get dissapointed so fast, there is something going on here you just have to troubleshoot, unless there is some incompatibility with your motherboard.
Computurd,
8 drives is double the amount of drives he's got to play with. (it doesn't matter if he's got 6Gb/s devices)
As I commented, if you look at the ATTO benchmark, the throughput is there, it's just that is located around 128KB-256KB and not where the typical benchmarks programs are measuring the throughput for sequential io's.
It should be easily reproducible using HD Tune using 128KB block size or iometer using the same block size.
well i can agree with that.... :) i just think there may be a little more that he could get from it with some tweaking, but you are correct, in the end it isnt going to give him what he is looking for, regardless of the amount of tweaking.
OK. Thanks for answers.
So i need LSI Megaraid series card for my needs? 9240-4i is the entry line. That should be able to give me the sequentials what i am looking for?
Im really dissapointed about lack of information on LSI 9211 series raid cards. I mean, why do they even make those HBA cards at sata3 if they are not even close to the sata3 speed requiements. And there is ZERO information about the limitation on their page.
Honestly. I starting to look some other brand now. This is discrace for LSI.
Silly me. I should have asked here, before i bought it. All info i found in forums and other pages suggested that 9211-4i is up to its task with 4x c300. Now i know, it clearly isnt!!!
I wouldn't give up so quickly. But it's 9260-4i not 9240.Quote:
Originally Posted by Kristoferr
9240-4i is the lowest cost of the series or im i wrong? 40 doesnt have on board cache. That should be the only difference. I can be wrong again tho...
Anyway. I already ordered 9260-4i because price difference wanst too much. So i cant go wrong this time :)
Just to make sure you understand - you ARE reaching 1.37GB/s reads at 128kb block size.Quote:
Originally Posted by Kristoferr
That's what I was trying to point out, I'm just not shure if he got it.Quote:
Originally Posted by One_Hertz
The 9211 is ultra-fast, just not for the sort of benchmarks he's been using.
In reality, it's almost as fast as the 9260 without FastPath.
i got that its fast at 128kb block size but thats pretty much it. It isnt fast enought while reading big files in sequental way? Or im getting it wrong again?
Anyway. I probably put the 9211 on to the HDDs or sell it in forums. Not too big loss since it was cheap.
One Hertz! nice to see you, seems like it's been a while! :)
really how often are you reading files much larger than 128kb? that card is a screamer man, ultra low latency and snappy as hell. i love it ! it is fast as heck, and very low overhead. in real world testing against the 9260 that i did (pre-fastpath) it was right there with the 9260 on game loads, etc. there is a point of diminishing returns of course. much over what you already have there isnt going to yield huge returns. ...unless you are doing tons of writing id stick with it, the cache on the 9260 is good for writing, of course.
Thats about it. Im doing it every day. Thats why im developing raid in the first place. Otherwise i would stick to 1 SSD only because "in the real world" theres not much difference between one good SSD and SSDs in raid 0.Quote:
Originally Posted by Computurd
I am sports market analyst in real world and im working files with a size of 1GB+ daily. :)