Mad222 - Core i7 3960X @ 5400MHz - 55334 marks PCMark 2005

[Guest John Lam]

SSD Power !!

[Dinos111]

Congratulations on the result, but! Can you explain that with this small controller took out 1259 General Usage & 1703 Virus Scan.

I mention this because neither Areca 1880 does not make this result with eight SSD.

Edited April 15, 2012 by Dinos111

[I.M.O.G.]

55K is broken. Nice result!

 

@Dinos: 1703 virus scan is easy with onboard raid. I've seen 2K virus scan with onboard, depends on mobo and RST. Virus scan is not exceptional, just very good.

 

The general usage is 400 higher than the best gen usage on hwbot! Areca 18XX and Revo 3 X2 480 have done 900 gen usage... No known setup has been tested at 1300 gen usage before.

[Guest John Lam]

@Dinos: This is Fusion-MPT 2.0 IO Controller , PCI-E 2.0 x8 , 8 SATA 6GBps , Max 4000MB/s Bandwidth and Max 320,000 IOPS in 4K Read !! easy to have 2K virus scan and over 2K gen usage ,setting and optimization is very important for RAID 0 SSD Array.

 

 

[I.M.O.G.]

The Areca 1882IX is specced at 4700MB/s bandwidth. It also supports 4GB DDR3 1333MHz ram for cache. The onboard ram cache is what yields high general usage performance... With 3x maxiops or 5x maxiops, it yields 800-900MB/s Gen Usage if configured correctly. Gen Usage does not scale with number of SSDs between 3 and 5, in my experience. This can also be correlated with Stevero's 1880 using 15x hardware ramdrives, or mikecdm running an older areca with 2GB onboard cache (slower ddr) - these sorts of cards can push 600-900 gen usage due to caching, storage disks matter very little.

 

What is housing the Fusion-MPT 2.0 IO Controller? Is this an LSI card? I would love to see some ATTO/Crystaldiskmark/ASSSD screens from this puppy.

 

This is all I have handy for the Areca in 3x RAID0:

Edited April 15, 2012 by I.M.O.G.

[Stelaras]

Hi Mad222 ,

 

Please forgive me for questioning your great result ...

 

From the picture , i can understand that this raid controller is an LSI 9211-8i .

 

I can't get it how one old raid controller from 2009 can score better than one highend 1880X which has 4GB+ of cache . I'm confused ...

 

Even the raid menu is the same with the 9211-8i . Also the specs that HCEPC wrote match exactly.

 

 

It's a value controller priced around $250 . There are threads from many forums about this controller performance .

 

http://www.xtremesystems.org/forums/showthread.php?241712-LSI-9211-8i-versus-LSI-9260-8i-versus-Areca-1231ML-2G

 

http://www.xtremesystems.org/forums/showthread.php?240574-LSI-9211-8i

 

 

It should be noted that this controller doesn't have onboard cache , so it doesn't support caching at all . Huge disadvantage over the other raid controllers .

 

http://www.lsi.com/downloads/Public/Host%20Bus%20Adapters/LSI_SAS9211-8i_PB_032211.pdf

 

http://www.lsi.com/downloads/Public/Obsolete/Obsolete%20Common%20Files/fusion.pdf

 

I would like to make the same request as I.M.O.G. . Can you please show as an as-ssd bench of your array ???

[TeamAU]

Wow crazy good general access!!

[Gluvocio]

very nice mad222, crazy storage and GDDR scores ... WR is your... for now

[SteveRo]

I played with a 9211 for some time and even with 8xR0 acards I was never able to get these kind of scores out of it. - could be I wasn't set up correctly?

 

[Guest John Lam]

http://www.hwbot.org/submission/2274630_hkepc_as_ssd_benchmark_ddr2_ramdisk_317_points?recalculate=true

 

 

Edited April 16, 2012 by HKEPC
add photo

[Guest John Lam]

http://www.hwbot.org/submission/2274639_hkepc_as_ssd_benchmark_ddr2_ramdisk_2463_points

 

 

Edited April 16, 2012 by HKEPC

[Tiltevros]

Really??????????????? Stop using Windows Compresion..... Its not legal its same as Dynamic Drive..... We re not fools.... Storage Sector feeds our home.... so plz dont post CPU/Ram sencitive caching things...

LSI 9260 is limited to 76.000 IOPS in IR controler by its own it only perform more than that in Dynamic or IT frimware... whitch is the same us ES cpu's or custom Frimware... Do u want really to start doing this on WR's?

[Mad222]

Guys, thx for all your concern

 

Please disregard the above ATTO/ASSSD tests provided by one of our members who actually wanted to prove its’ inaccuracy.

 

In fact, Windows Compression can’t enhance the efficiency of PCMark HDD

 

Screen Cap#1 No Windows Compression is used

http://i1140.photobucket.com/albums/n568/madtse/NoCompress.png

 

Screen Cap#2 Windows Compression is used

http://i1140.photobucket.com/albums/n568/madtse/Compressed.png

 

http://i1140.photobucket.com/albums/n568/madtse/research.png

 

http://i1140.photobucket.com/albums/n568/madtse/research-2.png

 

 

Yes, increased efficiency can be found in ATTO/ASSSD, but NOT in CrystalDiskMark as diff. testings are conducted. ATTO & ASSSD are OS based & CrystalDiskMark is I/O based

 

Futuremark PCMARK 05 HDD Tests are based on RankDisk developed by Intel. It records the trace of disk I/O activity during benchmarking and bypasses the file system and the operating system's cache. If we compress the data, tMB/s and IOPS will be lower and PC Mark HDD score will decrease accordingly

 

Our Lab has been conducting such study in the last few months & concluded some very interesting findings. As professional Overclockers, we hope we could look into things more in-depth.