What do your hard disks score on xbench?

[msmth928]

Download xbench (http://www.xbench.com) (from the top right) install and run. Uncheck everything bar the hard disk and post your scores :apple:

I have 2 500GB 16MB Cache Seagates that both average at around 85
I have another 250GB 16MB Seagate that averages about 52 (!)
The 640GB Hitachi that came with this Mac Pro averages at 72

I ran the test 3 times for each.

[Andrew Henry]

I'd tell you, but for some reason when I try to run the test on any of my 1TB WD Caviar Black drives it gives me an error! :(

:mad:

[voyagerd]

500GB Seagate Momentus 7200.4 in my MBP with SATA 150: 48.82

[Tesselator]

Three HD154UI Samsung F2 EcoGreen drives in a RAID0:
Results 176.78
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 12288 MB
Model MacPro1,1
Drive Type Macintosh RB2 3-Drive RAID0 4.5 TB

Disk Test 176.78
Sequential 175.77
Uncached Write 443.19 272.11 MB/sec [4K blocks]
Uncached Write 479.91 271.53 MB/sec [256K blocks]
Uncached Read 60.62 17.74 MB/sec [4K blocks]
Uncached Read 520.48 261.59 MB/sec [256K blocks]
Random 177.80
Uncached Write 73.69 7.80 MB/sec [4K blocks]
Uncached Write 460.97 147.57 MB/sec [256K blocks]
Uncached Read 306.47 2.17 MB/sec [4K blocks]
Uncached Read 286.15 53.10 MB/sec [256K blocks]


1TG Western Digital "green" drive:Results 75.72
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 12288 MB
Model MacPro1,1
Drive Type WDC WD10EACS-00ZJB0

Disk Test 75.72
Sequential 108.37
Uncached Write 96.34 59.15 MB/sec [4K blocks]
Uncached Write 118.58 67.09 MB/sec [256K blocks]
Uncached Read 89.06 26.06 MB/sec [4K blocks]
Uncached Read 145.55 73.15 MB/sec [256K blocks]
Random 58.19
Uncached Write 22.65 2.40 MB/sec [4K blocks]
Uncached Write 187.66 60.08 MB/sec [256K blocks]
Uncached Read 82.74 0.59 MB/sec [4K blocks]
Uncached Read 139.39 25.86 MB/sec [256K blocks]


Here's one from someone's Intel X25-M 80gb 2-drive RAID 0 (http://www.macintouch.com/readerreports/benchmarking/topic4490.html):Results 361.12
System Info
Xbench Version 1.3
System Version 10.5.6 (9G55)
Physical RAM 10240 MB
Model MacPro3,1
Drive Type System - X25-M 80gb 2-drive RAID0

Disk Test 361.12
Sequential 232.24
Uncached Write 236.81 145.40 MB/sec [4K blocks]
Uncached Write 206.42 116.79 MB/sec [256K blocks]
Uncached Read 143.56 42.01 MB/sec [4K blocks]
Uncached Read 840.16 422.26 MB/sec [256K blocks]

Random 811.32
Uncached Write 876.30 92.77 MB/sec [4K blocks]
Uncached Write 364.56 116.71 MB/sec [256K blocks]
Uncached Read 2225.26 15.77 MB/sec [4K blocks]
Uncached Read 1675.97 310.99 MB/sec [256K blocks]


Here's my RAMDisk:
Results 906.19
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 12288 MB
Model MacPro1,1
Drive Type Apple read/write RAM-Drive

Disk Test 906.19
Sequential 553.47
Uncached Write 415.27 254.97 MB/sec [4K blocks]
Uncached Write 1242.03 702.74 MB/sec [256K blocks]
Uncached Read 282.08 82.55 MB/sec [4K blocks]
Uncached Read 2132.99 1072.02 MB/sec [256K blocks]
Random 2498.42
Uncached Write 1195.78 126.59 MB/sec [4K blocks]
Uncached Write 2303.16 737.33 MB/sec [256K blocks]
Uncached Read 8109.22 57.46 MB/sec [4K blocks]
Uncached Read 4825.48 895.40 MB/sec [256K blocks]

[Cynicalone]

Bay 1 & 2 is setup in a RAID 1 with 2 WD 1TB Blacks. Bay 3 is a 1TB WD Black setup as JBOD. Bay 4 is so overloaded with crap right now I didn't even bother to test it. :)

[aibo]

here's mine with a single X25-M as boot drive full results
(http://db.xbench.com/merge.xhtml?doc1=349285)
Disk Test
267.07

[hehejames]

*** Two WD 1TB Black setup as RAID 0 ***

Results 168.19
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 16384 MB
Model MacPro3,1
Drive Type Me
Disk Test 168.19
Sequential 206.01
Uncached Write 320.12 196.55 MB/sec [4K blocks]
Uncached Write 309.90 175.34 MB/sec [256K blocks]
Uncached Read 95.45 27.93 MB/sec [4K blocks]
Uncached Read 386.20 194.10 MB/sec [256K blocks]
Random 142.09
Uncached Write 58.82 6.23 MB/sec [4K blocks]
Uncached Write 769.30 246.28 MB/sec [256K blocks]
Uncached Read 176.15 1.25 MB/sec [4K blocks]
Uncached Read 239.75 44.49 MB/sec [256K blocks]



*** WD 500GB SE16 ***

Results 84.74
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 16384 MB
Model MacPro3,1
Drive Type WDC WD5000AAKS-41YGA1
Disk Test 84.74
Sequential 129.24
Uncached Write 135.04 82.92 MB/sec [4K blocks]
Uncached Write 133.91 75.76 MB/sec [256K blocks]
Uncached Read 102.86 30.10 MB/sec [4K blocks]
Uncached Read 157.35 79.08 MB/sec [256K blocks]
Random 63.04
Uncached Write 24.03 2.54 MB/sec [4K blocks]
Uncached Write 202.91 64.96 MB/sec [256K blocks]
Uncached Read 95.75 0.68 MB/sec [4K blocks]
Uncached Read 154.46 28.66 MB/sec [256K blocks]

[VirtualRain]

Here's my pair of X25-M's in RAID0... :D

Results 438.45
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 6144 MB
Model MacPro4,1
Drive Type Intel SSD RAID0
Disk Test 438.45
Sequential 286.82
Uncached Write 276.71 169.90 MB/sec [4K blocks]
Uncached Write 253.02 143.16 MB/sec [256K blocks]
Uncached Read 189.87 55.57 MB/sec [4K blocks]
Uncached Read 898.12 451.39 MB/sec [256K blocks]
Random 930.26
Uncached Write 1095.54 115.98 MB/sec [4K blocks]
Uncached Write 417.89 133.78 MB/sec [256K blocks]
Uncached Read 2346.21 16.63 MB/sec [4K blocks]
Uncached Read 1760.86 326.74 MB/sec [256K blocks]



Here's a single 1TB WD Black (stock drive that came with my Mac Pro)...

Results 98.60
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 6144 MB
Model MacPro4,1
Drive Type WDC WD1001FALS-41K1B0
Disk Test 98.60
Sequential 168.08
Uncached Write 175.09 107.50 MB/sec [4K blocks]
Uncached Write 172.29 97.48 MB/sec [256K blocks]
Uncached Read 132.69 38.83 MB/sec [4K blocks]
Uncached Read 210.66 105.88 MB/sec [256K blocks]
Random 69.77
Uncached Write 24.85 2.63 MB/sec [4K blocks]
Uncached Write 318.09 101.83 MB/sec [256K blocks]
Uncached Read 115.63 0.82 MB/sec [4K blocks]
Uncached Read 188.95 35.06 MB/sec [256K blocks]

[VirtualRain]

Bay 1 & 2 is setup in a RAID 1 with 2 WD 1TB Blacks. Bay 3 is a 1TB WD Black setup as JBOD. Bay 4 is so overloaded with crap right now I didn't even bother to test it. :)

Interesting, my Random Write on a WD 1TB Black is 101MB/s, while yours is 24MB/s. :confused:

At any rate, assuming Diskbench is any good, you can see the impact on write performance running RAID1 in your case. There doesn't appear to be any improvement in read performance so I guess we can conclude that OSX software RAID does not use read optimization on RAID1 arrays.

[VirtualRain]

For those following Tesselator's and my debate on the merits of SSD's here's a quick comparison of his 3-drive RAID0 of hard drives to my 2-drive RAID0 of SSD's...

Sequential Magnetic SSD Difference
Uncached Write 272.11 169.90 -38%
Uncached Write 271.53 143.16 -47%
Uncached Read 17.74 55.57 +313%
Uncached Read 261.59 451.39 +173%
Random
Uncached Write 7.80 115.98 +1,487%
Uncached Write 147.57 133.78 -9%
Uncached Read 2.17 16.63 +766%
Uncached Read 53.10 326.74 +615%

While SSD's suck in $/GB, they definitely perform on read operations like no-body's business :)

[Cynicalone]

Interesting, my Random Write on a WD 1TB Black is 101MB/s, while yours is 24MB/s. :confused:

At any rate, assuming Diskbench is any good, you can see the impact on write performance running RAID1 in your case. There doesn't appear to be any improvement in read performance so I guess we can conclude that OSX software RAID does not use read optimization on RAID1 arrays.

That is with the Apple RAID card. I know slow, so slow.

[pustiu]

here's mine with a single X25-M as boot drive

[Decrepit]

Mac Mini G4 with 7200 rpm external G-Drive running off of FW400 connector...

Score: 33.44

My internal drive won't even run the test. Very odd.

[VirtualRain]

here's mine with a single X25-M as boot drive

Interesting... Comparing this to my Raid0 results all the tests scale as expected except the 4k reads. Curious.

[StanD]

My Velociraptors on a Highpoint RocketRAID 4310:

[Tesselator]

*** Two WD 1TB Black setup as RAID 0 ***

Results 168.19
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 16384 MB
Model MacPro3,1
Drive Type Me
Disk Test 168.19
Sequential 206.01
Uncached Write 320.12 196.55 MB/sec [4K blocks]
Uncached Write 309.90 175.34 MB/sec [256K blocks]
Uncached Read 95.45 27.93 MB/sec [4K blocks]
Uncached Read 386.20 194.10 MB/sec [256K blocks]
Random 142.09
Uncached Write 58.82 6.23 MB/sec [4K blocks]
Uncached Write 769.30 246.28 MB/sec [256K blocks]
Uncached Read 176.15 1.25 MB/sec [4K blocks]
Uncached Read 239.75 44.49 MB/sec [256K blocks]



For those following Tesselator's and my debate on the merits of SSD's here's a quick comparison of his 3-drive RAID0 of hard drives to my 2-drive RAID0 of SSD's...

Sequential Magnetic SSD Difference
Uncached Write 272.11 169.90 -38%
Uncached Write 271.53 143.16 -47%
Uncached Read 17.74 55.57 +313%
Uncached Read 261.59 451.39 +173%
Random
Uncached Write 7.80 115.98 +1,487%
Uncached Write 147.57 133.78 -9%
Uncached Read 2.17 16.63 +766%
Uncached Read 53.10 326.74 +615%

While SSD's suck in $/GB, they definitely perform on read operations like no-body's business :)

Not really a good comparrison tho. As I keep reminding everyone my 3-Drive RAID is composed of F2 EcoGREEN drives. Look at the 2-drive Black above. That scored 170 (about the same as my 3 green drives) and with a 3rd drive will probably be 250 ~ 275. That will be the same as aibo's results here:

here's mine with a single X25-M as boot drive full results
(http://db.xbench.com/merge.xhtml?doc1=349285)
Disk Test
267.07

Also you kinda left out the fact that your scores here are from two SSDs in RAID0 @ $700. Our discussion was over a $300 budget. One SSD compared to a 3-drive RAID0 of rotational media. :p

And hehejames's black drives aren't even 500GB Platters. Wait 4 to 6 months till 500GB platters ship in non-green type units and likely a good 3-drive RAID will kick butt. :D But anyway I think these scores here show very clearly the points I was making in the other thread. The two are about the same but in the case of the RAID you additionally get a capacity that you can really use. Both are good to have over a single drive! Both are fast! And if you're rich enough to go for an SSD in the 1st place just RAID0 three of the 250GB SSD units and no one will be able to touch you in any benchmark. :p Of course it'll be $2,000 but hey... :)

.

[aibo]

And aibo's black drives aren't even 500GB Platters. Wait 4 to 6 months till 500GB platters ship in non-green type units and likely a good 3-drive RAID will kick butt. :D But anyway I think these scores here show very clearly the points I was making in the other thread. The two are about the same but in the case of the RAID you additionally get a capacity that you can really use. Both are good to have over a single drive! Both are fast! And if you're rich enough to go for an SSD in the 1st place just RAID0 three of the 250GB SSD units and no one will be able to touch you in any benchmark. :p Of course it'll be $2,000 but hey... :)

.
I don't have caviar black drives... and I don't see any need to ever go RAID again, after experiencing SSD. One single $320 SSD is outperforming your 3-drive RAID setup.

[Shake 'n' Bake]

My mid-'07 mini's internal, called "Macintosh HD," scored a whopping 24.96! That's with the stock 120 GB HDD.

The external, connected via FW 400, scored 41.38. It is a Maxtor One Touch 4 Plus.

My PowerMac G4 DA's internal, called "Alpha," scored 34.93. "Alpha" is the stock 60 GB HDD. The other internal, called "Beta," scored 28.41. "Beta" is a 40 GB Maxtor pulled from a Dell Dimension 4300S.

[Tesselator]

I don't have caviar black drives...

Yeah, sorry I meant hehejames's. :D

I'll fix it in my post tho thanks!


.

[Flash SWT]

Two Seagate 750GB ST3750330NS in a RAID 1 mirror:
Results 69.43
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 8192 MB
Model MacPro4,1
Drive Type Mac RAID 1
Disk Test 69.43
Sequential 120.21
Uncached Write 157.54 96.73 MB/sec [4K blocks]
Uncached Write 154.30 87.30 MB/sec [256K blocks]
Uncached Read 80.77 23.64 MB/sec [4K blocks]
Uncached Read 123.99 62.31 MB/sec [256K blocks]
Random 48.81
Uncached Write 15.70 1.66 MB/sec [4K blocks]
Uncached Write 272.57 87.26 MB/sec [256K blocks]
Uncached Read 112.40 0.80 MB/sec [4K blocks]
Uncached Read 175.87 32.63 MB/sec [256K blocks]


Stock Western Digital 640GB Blue:
Results 75.70
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 8192 MB
Model MacPro4,1
Drive Type WDC WD6400AAKS-41H2B0
Disk Test 75.70
Sequential 148.98
Uncached Write 160.87 98.77 MB/sec [4K blocks]
Uncached Write 159.83 90.43 MB/sec [256K blocks]
Uncached Read 108.70 31.81 MB/sec [4K blocks]
Uncached Read 193.16 97.08 MB/sec [256K blocks]
Random 50.75
Uncached Write 17.25 1.83 MB/sec [4K blocks]
Uncached Write 223.91 71.68 MB/sec [256K blocks]
Uncached Read 96.07 0.68 MB/sec [4K blocks]
Uncached Read 167.32 31.05 MB/sec [256K blocks]


Western Digital 1.5 TB WD15EADS gives an error and won't test.


.

[vicentk]

http://i104.photobucket.com/albums/m166/vicentk/Picture1-6.png

4*Raid 0 WD6400AALS

[Tesselator]

Nice! And those are the 640GB WD Blue drives too right?

[VirtualRain]

So, why do conventional hard drives suck so bad at 4K random reads? I'm assuming it's the access times?

[lemonade-maker]

Results 181.85
System Info
Xbench Version 1.3
System Version 10.x
Physical RAM 16384 MB
Model MacPro4,1
Drive Type Earth
Disk Test 181.85
Sequential 269.43
Uncached Write 495.98 304.53 MB/sec [4K blocks]
Uncached Write 604.42 341.98 MB/sec [256K blocks]
Uncached Read 102.31 29.94 MB/sec [4K blocks]
Uncached Read 713.49 358.59 MB/sec [256K blocks]
Random 137.24
Uncached Write 53.73 5.69 MB/sec [4K blocks]
Uncached Write 369.70 118.35 MB/sec [256K blocks]
Uncached Read 245.59 1.74 MB/sec [4K blocks]
Uncached Read 266.12 49.38 MB/sec [256K blocks]

[Tesselator]

So, why do conventional hard drives suck so bad at 4K random reads? I'm assuming it's the access times?

Yeah, I think so too. Access as in both head seek and rotation.

[hyram]

Results 617.08
System Info
Xbench Version 1.3
System Version 10.4.11 (8S2167)
Physical RAM 8192 MB
Model MacPro1,1
Drive Type HPT DISK 0_0
Disk Test 617.08
Sequential 383.30
Uncached Write 212.63 130.55 MB/sec [4K blocks]
Uncached Write 1261.81 713.93 MB/sec [256K blocks]
Uncached Read 228.65 66.91 MB/sec [4K blocks]
Uncached Read 1764.48 886.81 MB/sec [256K blocks]
Random 1581.94
Uncached Write 571.29 60.48 MB/sec [4K blocks]
Uncached Write 2206.56 706.40 MB/sec [256K blocks]
Uncached Read 8474.08 60.05 MB/sec [4K blocks]
Uncached Read 4832.41 896.69 MB/sec [256K blocks]

[Tesselator]

Results 617.08
System Info
Xbench Version 1.3
System Version 10.4.11 (8S2167)
Physical RAM 8192 MB
Model MacPro1,1
Drive Type HPT DISK 0_0
Disk Test 617.08
Sequential 383.30
Uncached Write 212.63 130.55 MB/sec [4K blocks]
Uncached Write 1261.81 713.93 MB/sec [256K blocks]
Uncached Read 228.65 66.91 MB/sec [4K blocks]
Uncached Read 1764.48 886.81 MB/sec [256K blocks]
Random 1581.94
Uncached Write 571.29 60.48 MB/sec [4K blocks]
Uncached Write 2206.56 706.40 MB/sec [256K blocks]
Uncached Read 8474.08 60.05 MB/sec [4K blocks]
Uncached Read 4832.41 896.69 MB/sec [256K blocks]

Wow! What a difference a controller makes! Sweet!

[sparkie7]

Here's my RAMDisk:
Results 906.19
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 12288 MB
Model MacPro1,1
Drive Type Apple read/write RAM-Drive

Disk Test 906.19
Sequential 553.47
Uncached Write 415.27 254.97 MB/sec [4K blocks]
Uncached Write 1242.03 702.74 MB/sec [256K blocks]
Uncached Read 282.08 82.55 MB/sec [4K blocks]
Uncached Read 2132.99 1072.02 MB/sec [256K blocks]
Random 2498.42
Uncached Write 1195.78 126.59 MB/sec [4K blocks]
Uncached Write 2303.16 737.33 MB/sec [256K blocks]
Uncached Read 8109.22 57.46 MB/sec [4K blocks]
Uncached Read 4825.48 895.40 MB/sec [256K blocks]

what are using to make this RAM disk. Terminal or an app like Make RAM Disk?

Also how are you using the RAM disk

I posted a thread a while back about the benefits of using a RAM disk, but most people seemed to think it was a waste of time as OSX utilises unused RAM effieciently

[sparkie7]

Results 617.08
System Info
Xbench Version 1.3
System Version 10.4.11 (8S2167)
Physical RAM 8192 MB
Model MacPro1,1
Drive Type HPT DISK 0_0
Disk Test 617.08
Sequential 383.30
Uncached Write 212.63 130.55 MB/sec [4K blocks]
Uncached Write 1261.81 713.93 MB/sec [256K blocks]
Uncached Read 228.65 66.91 MB/sec [4K blocks]
Uncached Read 1764.48 886.81 MB/sec [256K blocks]
Random 1581.94
Uncached Write 571.29 60.48 MB/sec [4K blocks]
Uncached Write 2206.56 706.40 MB/sec [256K blocks]
Uncached Read 8474.08 60.05 MB/sec [4K blocks]
Uncached Read 4832.41 896.69 MB/sec [256K blocks]


That is SICK. TOTALLY SiCK!

Give us more!!

However you're failure rate is exponential :eek:

[Tesselator]

what are using to make this RAM disk. Terminal or an app like Make RAM Disk?

Also how are you using the RAM disk

I posted a thread a while back about the benefits of using a RAM disk, but most people seemed to think it was a waste of time as OSX utilises unused RAM effieciently

I'm using Esperance DV 2.3.2 that I got from here: http://www.macupdate.com/info.php/id/16518

Someone told me that PS CS4 Extended loaded in 1 second from an SSD and I didn't believe it. So I downloaded this app, created a 2GB drive, installed CS4 on it and gave it a try with something that is 4 times faster than an SSD and is the more's law equivalent. Needless to say loading a properly configured PS CS4 in anything close to 1sec. is an impossibility but that's why I have it and what I'm currently using it for. PS is still installed on it after several reboots and I was even able to shrink it's size down closer to PS's needs - which for me is 800MB. :)

As for the naysayers saying RAM Disks such as this aren't fast just look no further than the benchmark. How useful it is depends on you of course. This particular one only supports sizes up to 2GB I think so that's a little limited. There may be other warez or methods to make larger sizes I dunno. Have you tried the shell script method? How's that? And lastly they may be right about the efficiency. The way that OS X handles the system cache it will at first anyway essentially be in ram twice. And maybe three times (depending how you look at it) at least temporarily.

Anyway, it's your system. Do what you want and don't let others talk you out of it. You'll know soon enough if it's something good for you. If it's not just remove it. :)

However you're failure rate is exponential :eek:

That's a bit of a myth. You know?

.

[hyram]

That is SICK. TOTALLY SiCK!

Give us more!!

However you're failure rate is exponential :eek:


I know... it's only temporary. I'll pick up 2 more drives this week and convert it to RAID 5. I'll post those results when I have them

hyram

[VirtualRain]

Results 617.08
System Info
Xbench Version 1.3
System Version 10.4.11 (8S2167)
Physical RAM 8192 MB
Model MacPro1,1
Drive Type HPT DISK 0_0
Disk Test 617.08
Sequential 383.30
Uncached Write 212.63 130.55 MB/sec [4K blocks]
Uncached Write 1261.81 713.93 MB/sec [256K blocks]
Uncached Read 228.65 66.91 MB/sec [4K blocks]
Uncached Read 1764.48 886.81 MB/sec [256K blocks]
Random 1581.94
Uncached Write 571.29 60.48 MB/sec [4K blocks]
Uncached Write 2206.56 706.40 MB/sec [256K blocks]
Uncached Read 8474.08 60.05 MB/sec [4K blocks]
Uncached Read 4832.41 896.69 MB/sec [256K blocks]

There must be some cache effect at work here... for example, a single WD Black will score under 1MB/s on random 4k reads, but here the array is scoring 60MB/s which is not just 6 times a single drive but 60 times! :confused:

There are some cases where Xbench seems to fall short of being an accurate benchmarking tool and this is perhaps one. At any rate, I expect 6 WD Blacks in HW RAID0 would rock just as this suggests! :eek:

[Tesselator]

There must be some cache effect at work here... for example, a single WD Black will score under 1MB/s on random 4k reads, but here the array is scoring 60MB/s which is not just 6 times a single drive but 60 times! :confused:

That's classical RAID0. As the number of volumes increase access times decrease and at the same time throughput is increased, The affects compound and scores like this are achieved. You can see this more clearly if you look at detailed samplings from 1, 2, 3, 4, 5, and then 6 drive RAID's benchmarks and compare them. I've been up to 4 on the MP. He's got a card so he can go to 6 (or 8?) if he wants. :) There's documents on-line that take it up to 8. Very informative. The effects this has on the user's experience are immediately apparent.

[nanofrog]

That's classical RAID0. As the number of volumes increase access times decrease and at the same time throughput is increased, The affects compound and scores like this are achieved. You can see this more clearly if you look at detailed samplings from 1, 2, 3, 4, 5, and then 6 drive RAID's benchmarks and compare them. I've been up to 4 on the MP. He's got a card so he can go to 6 (or 8?) if he wants. :) There's documents on-line that take it up to 8. Very informative. The effects this has on the user's experience are immediately apparent.
Yup. And once a user figures out the best stripe size for their specific usage, it gets better. ;)

[Tesselator]

Yup. And once a user figures out the best stripe size for their specific usage, it gets better. ;)

Yup, and more fine tuning is available by tweaking the RAID Block Size.

[nanofrog]

Yup, and more fine tuning is available by tweaking the RAID Block Size.
I'm used to stripe size = block size, chunk size, stripe length or granularity

Are you referring to OS block size? Something else? :confused:

[Tesselator]

It's a option setting in Apple's Software Raid Scheme:

http://images.macworld.com/images/legacy/2007/05/images/content/blocksize.jpg

I believe it's the chunk size. :)

[iGary]

WD Velociraptor:

Wish I knew more about setting up RAIDs, but it's quickie enough for me. :)

[msmth928]

How much can I gain if I put my two 500GB seagates into Raid0 ? (both score 85).

I can then get a 1GB drive for time machine :D

[hehejames]

How much can I gain if I put my two 500GB seagates into Raid0 ? (both score 85).

Most likely in the range of 140-160. See additional results from page 1.

[msmth928]

Thanks, that looks great! Is that with a raid card? If so which one?

Also, can you notice any difference in everyday use?

[VirtualRain]

Thanks, that looks great! Is that with a raid card? If so which one?

Also, can you notice any difference in everyday use?

You don't need a RAID card for RAID0. OSX has great software RAID support that scales well.

As for noticing a difference... this is a subject of great debate. In my opinion, it will make a subtle difference... more obvious with reading/writing large files. However, the greatest benefit is gaining a larger effective volume from smaller disks translating into less juggling of storage tasks between drives.

[hehejames]

Thanks, that looks great! Is that with a raid card? If so which one?

No. It's using Apple's disk utility. As an FYI, you can use SoftRAID and many seem to like it very much. For more info, you can refer to Lloyd's guide http://www.macperformanceguide.com/Storage-RAID.html

can you notice any difference in everyday use?

I'm not a Graphics/Video person (plain business user), but for me I do see the difference moving/opening large files back and forth and running certain apps (hosting VM image files and running my VM's off the RAID 0 drive).

The different would be much more significant if you were rendering and/or dealing with massive PSD files and etc.

[UltraNEO*]

Just be aware.

Setting up your system with a RAID0 (whether Soft or Hard RAID) doubles the chances of data loss. Simply because there's twice as many things which can go wrong, double the chances of something failing. Remember, if one of the HD's fail, you'll automatically lose the data on BOTH hard drives.

Therefore, RAID0 in my opinion is best used as a fast scratch disk, where the data isn't so important, as the actual file is stored on another drive.

For safety I would strongly advise you to make backups of everything, regularly. At least once a week.

[msmth928]

Thanks for the info all. :apple:

I'd only do it after getting another HD for time machine back ups. At the mo one of the 500GB drives is being used for time machine, but with HDs so cheap these days I might just get a 1TB drive and put these two in Raid0 as they are a pair.

[nanofrog]

It's a option setting in Apple's Software Raid Scheme:

http://images.macworld.com/images/legacy/2007/05/images/content/blocksize.jpg

I believe it's the chunk size. :)
It's the stripe size, just worded a little differently. :)

[Tesselator]

Also, can you notice any difference in everyday use?

Yes, you will immediately notice the difference. Your mac will over all feel as tho it got a speed injection. No one can not notice the difference unless they're asleep at the helm. At 3-Drive RAID0 the difference is in the OMG category. At 4 and above you will be picking up your jaw until you get used to it and start taking it for granted.

[hyram]

You guys caught me at the right time as I'm in process of building and testing my RAID. Heres another test, but this time with 8x1TB TD Blacks in RAID0. As expected, a 25% improvement over my 6 disk configuration.

Results 782.31
System Info
Xbench Version 1.3
System Version 10.4.11 (8S2167)
Physical RAM 8192 MB
Model MacPro1,1
Drive Type HPT DISK 0_0
Disk Test 782.31
Sequential 556.89
Uncached Write 523.52 321.43 MB/sec [4K blocks]
Uncached Write 1351.34 764.59 MB/sec [256K blocks]
Uncached Read 250.32 73.26 MB/sec [4K blocks]
Uncached Read 1859.92 934.78 MB/sec [256K blocks]
Random 1314.34
Uncached Write 462.74 48.99 MB/sec [4K blocks]
Uncached Write 1830.93 586.15 MB/sec [256K blocks]
Uncached Read 8434.01 59.77 MB/sec [4K blocks]
Uncached Read 4596.27 852.87 MB/sec [256K blocks]

I ran the same test again, but this time with the cache turned off:

Results 237.62
System Info
Xbench Version 1.3
System Version 10.4.11 (8S2167)
Physical RAM 8192 MB
Model MacPro1,1
Drive Type HPT DISK 0_0
Disk Test 237.62
Sequential 164.49
Uncached Write 606.62 372.45 MB/sec [4K blocks]
Uncached Write 322.96 182.73 MB/sec [256K blocks]
Uncached Read 57.56 16.85 MB/sec [4K blocks]
Uncached Read 454.58 228.47 MB/sec [256K blocks]
Random 427.82
Uncached Write 321.00 33.98 MB/sec [4K blocks]
Uncached Write 560.49 179.43 MB/sec [256K blocks]
Uncached Read 455.11 3.23 MB/sec [4K blocks]
Uncached Read 443.85 82.36 MB/sec [256K blocks]

The affect of caching is significant.

Below is how it looks to Black Magic Speed test. As I don't really intend to run as RAID0 It's currently building RAID5. It'll take about a day, I'll post results when I have them.

hyram

[nanofrog]

The affect of caching is significant.
It is. ;) Especially when you have a card that can hold 2 or 4GB worth via a single DIMM/SO-DIMM. :D

[VirtualRain]

Random 427.82
Uncached Write 321.00 33.98 MB/sec [4K blocks]
Uncached Write 560.49 179.43 MB/sec [256K blocks]
Uncached Read 455.11 3.23 MB/sec [4K blocks]
Uncached Read 443.85 82.36 MB/sec [256K blocks]

The affect of caching is significant.


Yes, this is more consistent with what I would expect from a multi-drive stripe of magnetic disks on this benchmark.

[nightfly13]

Depressingly, the stock 250gb drive that shipped with my first gen 2.66 quad system (dual dual) comes up with a Disk Test of 33.41. Ick! This is my system drive! I've been itching to get a SSD in here because I've felt my hard drive speeds are lackluster - and now I see how appalling they really are! I also have some 500gb Samsung Spinpoints that only score 37-40 and the newest drive my modest collection, a year old 640gb Spinpoint scores best around 90.

I'm sure a SSD will help feel and performance (planning for the 80gb Intel) but is there anything I can do about my archival data drives?

Also does anyone know offhand how easy it is to utilize the SATA ports on the logicboard of the 1,1 Mac Pros? Any link would be appreciated - so I don't have to yank the 250gb out to install the SDD but just slam it in there somewhere.

[hyram]

Ok... here's the results for 8 x 1 TB WD Blacks in RAID 5 with a Highpoint 4322 controller:

Results 571.59
System Info
Xbench Version 1.3
System Version 10.4.11 (8S2167)
Physical RAM 8192 MB
Model MacPro1,1
Drive Type HPT DISK 0_0
Disk Test 571.59
Sequential 495.66
Uncached Write 532.11 326.71 MB/sec [4K blocks]
Uncached Write 1153.58 652.69 MB/sec [256K blocks]
Uncached Read 211.75 61.97 MB/sec [4K blocks]
Uncached Read 1663.13 835.88 MB/sec [256K blocks]
Random 675.01
Uncached Write 370.79 39.25 MB/sec [4K blocks]
Uncached Write 1086.92 347.96 MB/sec [256K blocks]
Uncached Read 1930.86 13.68 MB/sec [4K blocks]
Uncached Read 558.35 103.61 MB/sec [256K blocks]

Not as good as the 6 drives in RAID 0, but a whole lot more reliable.


Here's the results of Disktester area scan:

---------------- Averages for "RAID" (512MB/32MB, 10 iterations) -------------
Area (6.36TB) Write MB/sec Read MB/sec
0% 751 480
5% 758 474
10% 743 480
15% 747 480
20% 728 458
25% 733 467
30% 724 452
35% 703 449
40% 714 446
45% 705 449
50% 684 434
55% 692 423
60% 686 414
65% 648 404
70% 635 389
75% 622 376
80% 584 355
85% 573 352
90% 523 320
95% 501 304
100% 467 278
Average write speed across the volume: 663MB/sec
Average read speed across the volume: 414MB/sec

And below is the Black Magic Speed Test.

I think I can live with these numbers.

hyram

[Tesselator]

Ok... here's the results for 8 x 1 TB WD Blacks in RAID 5 with a Highpoint 4322 controller:

Here's the results of Disktester area scan:

---------------- Averages for "RAID" (512MB/32MB, 10 iterations) -------------
Area (6.36TB) Write MB/sec Read MB/sec
0% 751 480
5% 758 474
10% 743 480
15% 747 480
20% 728 458
25% 733 467
30% 724 452
35% 703 449
40% 714 446
45% 705 449
50% 684 434
55% 692 423
60% 686 414
65% 648 404
70% 635 389
75% 622 376
80% 584 355
85% 573 352
90% 523 320
95% 501 304
100% 467 278
Average write speed across the volume: 663MB/sec
Average read speed across the volume: 414MB/sec


Really? I remember my Level 5's being faster than this. This is not significantly different than my current RAID0 3-drive configuration (http://forums.macrumors.com/showpost.php?p=7673878&postcount=1). Hmm...

[nanofrog]

hyram, is the cache still switched off?

[hyram]

The cache is set to write back. Sector size is 512B, block size is 256k, NCQ is enabled INT 13 support is enabled (not sure about this one). These are the same settings I had when set as RAID 0.

I'm open to suggestions to try and improve this.

hyram

[nanofrog]

The cache is set to write back. Sector size is 512B, block size is 256k, NCQ is enabled INT 13 support is enabled (not sure about this one). These are the same settings I had when set as RAID 0.

I'm open to suggestions to try and improve this.

hyram
SAS uses TCQ, not NCQ. :eek: :(

So set NCQ = Disabled. You'll should see a difference. ;)
(BTW, don't try to enable TCQ either). :p

[Virtuoso]

Caldigit RAID5 with 4x WD RE2 drives vs Corsair SSD 256GB...

[hyram]

SAS uses TCQ, not NCQ. :eek: :(

So set NCQ = Disabled. You'll should see a difference. ;)
(BTW, don't try to enable TCQ either). :p

Thanks for the suggestion.. I'll give it a try as soon as I can. And as far as I know I can't enable TCQ.

hyram

[aibo]

If anyone wasn't already convinced SSD is the way to go, this thread should do it. People are needing 6+ drive RAIDs just to reach the performance level of a single Intel/OCZ SSD.

[Tesselator]

If anyone wasn't already convinced SSD is the way to go, this thread should do it. People are needing 6+ drive RAIDs just to reach the performance level of a single Intel/OCZ SSD.

:confused: :confused: :confused:


2-SSD RAID0 Test Score: ---------------- 438.45
Single SSD Disk Test Result: ------------ 267.07
Single SSD Disk Test Result: ------------ 249.52
8-Drive RAID0 Disk Test Result: -------- 782.31
6-Drive RAID0 Disk Test Results: ------- 617.08
4-Drive WD 640GB RAID Test Results: -- 238
4-Drive WD VR RAID Test Results: ------ 439.75
3-Drive GREEN RAID Test Results: ------ 176.78
2-Drive BLACK RAID Test Results: ------ 168.69


Q. Items B and C have similar results to what other items in the list?

Q. Which items in the list have better test results than either B or C?

[hyram]

"If anyone wasn't already convinced SSD is the way to go, this thread should do it. People are needing 6+ drive RAIDs just to reach the performance level of a single Intel/OCZ SSD."

Debatable… and only if speed is you're only criteria. I'm working with HD video, I have to trade off speed AND capacity. When file sizes approach 200-300 GB a moderate sized RAID fills up fast, a SSD is already full. And the $/GB is not even close… how much would I have to spend to get an SSD in the neighborhood of 7TB???

Don’t get me wrong… SSD has it’s place; I’m toying with the idea of an SSD for a boot/app drive. But I’ll probable wait for pricess to drop and capacity to increase a little more first.

But this RAID is for data only.

hyram

[VirtualRain]

I'm not sure Xbench adequately captures the impact that sub-millisecond access times have on every-day use.

I suspect the 4K Random read performance benchmark is the best measure of the impact that SSD's can have in most every-day applications... in some cases, an order of magnitude improvement. That's the only thing I can think of to make me such a fan-boy (besides the cool-aid that Intel keeps sending me! :p :D)

[Tesselator]

Hehe... well at least you're straight about it. That's kewl!

Do keep in mind tho that most of the accesses in a 3 or 4 drive RAID will be sub-millisecond too. Mine are. :)

[hyram]

SAS uses TCQ, not NCQ. :eek: :(

So set NCQ = Disabled. You'll should see a difference. ;)

It made absolutely no difference it NCQ is set or not.

I've been playing around with the test... various chunk sizes and file sizes, and I find this has the biggest impact. For example: iIf I set it to what was used in the AMUG review 128MB/10GB then I get almost identical performance to what AMUG reports. Difference is they are running Samsung's while I have the WD drives. Bt not a very big difference in the results. I'm going to try and do a more thurough study on this but it will have to wait a few days.

Tesselator, when I look at your results, they show fairly flat accross the whole drive surface. Most tests I've seen usually show some reduction in speed as the drive fills up. How are you results possible???

hyram

[nanofrog]

It made absolutely no difference it NCQ is set or not.

I've been playing around with the test... various chunk sizes and file sizes, and I find this has the biggest impact. For example: iIf I set it to what was used in the AMUG review 128MB/10GB then I get almost identical performance to what AMUG reports. Difference is they are running Samsung's while I have the WD drives. Bt not a very big difference in the results. I'm going to try and do a more thurough study on this but it will have to wait a few days.
It can on SAS cards. At least enough you can notice, and is supported by tests. It's not too much now on the Areca's I've used (since firmware 1.45). But as your's is a little different (HighPoint), I figured it was worth a shot. ;) :p

Stripe size can make a big difference, but had assumed you'd already settled on that (dialed it in for your usage/gotten advice elsewhere on it), since you hadn't asked. :eek: Oops. :o

[Tesselator]

Tesselator, when I look at your results, they show fairly flat accross the whole drive surface. Most tests I've seen usually show some reduction in speed as the drive fills up. How are you results possible???

hyram

I have no idea. It's just the way those particular drives profile. Keep in mind that these are GREEN drives and from what I read once the spindle speed is not only low 5900 RPM (not 5400 like some resale sites list it as) according to one Japanese source, it's variable speed - which sounds strange to me. So I wonder if those things have anything to do with it? Which benchmarks are you referring to specifically?

But if you look here: http://forums.macrumors.com/showpost.php?p=7673878&postcount=1 you will see the tapering off that you're talking about everywhere the drives aren't slamming up against some other limit.

[nanofrog]

Keep in mind that these are GREEN drives and from what I read once the spindle speed is not only low 5900 RPM (not 5400 like some resale sites list it as) according to one Japanese source, it's variable speed - which sounds strange to me.
This is what I'd expect for that drive. WD is doing the same thing to boost performance on their "Green" lines.

From what I've seen, they typically spin between 5400 and 7200rpm (getting faster as it gets to the inner tracks). :)

[hyram]

Mostly from AMUG reviews, but I've seen it on other sites as well. I uess I hadn't scrolled down enough on your graphs because the larger file sizes in your data do exhibit this as well.

I'm not sure I have much here to complain about, most of the time the rates at full are a little less than half the rate of empty drives. For mine the rates are a little more than half at full.

Write (MB/sec) Read (MB/sec)
706 689
695 678
677 659
668 647
643 624
621 603
585 564
544 522
501 487
442 426
370 359


This is with 10GB files in 128MB chunks.

It'll take me while,but I plan to do more thorough testing of the file/chunk size.

Mr. Frog... it didn't hurt to try it. I've just left it off as I can't see a difference with or without.

Thanks,

hyram

[nanofrog]

Mr. Frog... it didn't hurt to try it. I've just left it off as I can't see a difference with or without.

Thanks,

hyram
Mr.? :eek: Where? :p

Take it as a good thing. :D That means the firmware has gotten good enough that the difference has diminished, though not totally gone, as the processor is still SAS optimized. It took a while, but Areca seems to have gotten it sorted rather well with v1.45 (now at v1.46, but I've not run SATA on this revision). So perhaps the difference has dropped even further. :D :D

I wouldn't worry about it at any rate. :p

[Tesselator]

I'm not sure I have much here to complain about, most of the time the rates at full are a little less than half the rate of empty drives. For mine the rates are a little more than half at full.

Write (MB/sec) Read (MB/sec)
706 689
695 678
677 659
668 647
643 624
621 603
585 564
544 522
501 487
442 426
370 359

This is with 10GB files in 128MB chunks.

It'll take me while,but I plan to do more thorough testing of the file/chunk size.

Yeah, this is true. As a drive or raid set approaches 80% full performance begins to suffer a lot! I guess it's the combination of volume & file fragmentation along with likely having to work more with it's inner slower sectors.

The remedy these days is pretty straight forward. Use a big-ass RAID set and don't let it get more than 50% full. That way you're always on the cusp of it's high performance cylinder bands.

[Dr.Pants]

Coming in with a Hitachi Deskstar at 500GB capacity and 7200RPM, itssss...

DR.PANTS!!!

Disk Test 71.01
Sequential 108.33
Uncached Write 105.68 64.89 MB/sec [4K blocks]
Uncached Write 103.21 58.39 MB/sec [256K blocks]
Uncached Read 108.81 31.84 MB/sec [4K blocks]
Uncached Read 116.52 58.56 MB/sec [256K blocks]
Random 52.82
Uncached Write 20.32 2.15 MB/sec [4K blocks]
Uncached Write 111.03 35.55 MB/sec [256K blocks]
Uncached Read 95.88 0.68 MB/sec [4K blocks]
Uncached Read 141.37 26.23 MB/sec [256K blocks]

Anything look "off" on this to the initiated:confused:

[Virtuoso]

WD is doing the same thing to boost performance on their "Green" lines. From what I've seen, they typically spin between 5400 and 7200rpm (getting faster as it gets to the inner tracks). :)

I think this has been exposed as Marketing 'misdirection'. :rolleyes: The Green drives are a fixed speed, not a variable speed, of either 5400 and 7200, although apparently all the ones currently in the field run at 5400. From the SilentPC review:-

Most of a drive's power is consumed by the motor that spins the disk inside the drive. Reduce the speed of the disk, and you reduce the amount of power required. However, Western Digital doesn't want to say that they're selling 5,400 RPM drives — those became second class in the desktop market years ago. Instead, they rate the drive's speed as "IntelliPower" and take pains to emphasize that there are other factors that affect performance.

Western Digital has caught a lot of flak for withholding the rotation speed of the Green Power, especially when the product was first launched and the marketing material listed the rotation speed as 5,400-7,200 RPM. This led some to speculate that the rotation speed changed dynamically during use — which would have been an impressive engineering feat had it been true. The reality is revealed by a sentence that Western Digital added to the description of IntelliPower: "For each GreenPower™ drive model, WD may use a different, invariable RPM." In other words, Western Digital reserves the right to release both 5,400 RPM and 7,200 RPM drives under the Green Power name — without telling you which are which.

We were able to confirm that our 750 GB Green Power had a spindle speed of 5,400 RPM by doing frequency analysis on a sound recording of it. Why sound? Sound is vibration; the pitch of the sound corresponds to the frequency of the vibration. Hard drives vibrate at the speed of their motor, so they produce a noise at the same frequency as their rotation speed. Our sample had a sharp spike at exactly 90 Hz (cycles per second). Multiplying that number by 60 (to get cycles per minute) yielded a measured rotation speed of 5,400 RPM.

It's possible that other Green Power models use a higher spindle speed — but we doubt it. Storage Review tested the 1 TB version of the drive and determined that that model also spun at 5,400 RPM based on a calculation of the drive's latency compared to a previous Western Digital model. That leaves the 500 GB model — which Western Digital says is even lower power than the larger capacity versions. With the majority of the Green Power's efficiency advantage coming from its lower speed, it seems impossible for the 500 GB model to use a higher rotation speed. It's possible Western Digital intends to release a 7,200 RPM version at some point in the future.

[nanofrog]

I think this has been exposed as Marketing 'misdirection'. :rolleyes: The Green drives are a fixed speed, not a variable speed, of either 5400 and 7200, although apparently all the ones currently in the field run at 5400. From the SilentPC review:-
Nice catch. :D

I hadn't spotted this, but I guess I shouldn't be surprised either. :rolleyes: Given how drive makers determine capacity, and other values in the specifications. ;)

[Tesselator]

I don't think this changes anything. I think this is still just another person who doesn't actually know's opinion. I think it should be weighed as such too. It's for sure not even empirically definitive. For example he has to test the speed by recording vibrations. Yeah, but what was the drive doing at the time and did he try it while the drive was forced to access the inner most cylinders? The only thing in all this text that may not be wholly speculative is the mention of WD adding the "For each GreenPower™ drive model, WD may use a different, invariable RPM." text to their site - which I haven't seen yet and don't know the context. :)

So for me it could still likely be either way and no one knows what the spindle speed of the Samsung F2 EcoGreen HD154UI is yet. :p

[nanofrog]

I don't think this changes anything. I think this is still just another person who doesn't actually know's opinion. I think it should be weighed as such too. It's for sure not even empirically definitive. For example he has to test the speed by recording vibrations. Yeah, but what was the drive doing at the time and did he try it while the drive was forced to access the inner most cylinders? The only thing in all this text that may not be wholly speculative is the mention of WD adding the "For each GreenPower™ drive model, WD may use a different, invariable RPM." text to their site - which I haven't seen yet and don't know the context. :)

So for me it could still likely be either way and no one knows what the spindle speed of the Samsung F2 EcoGreen HD154UI is yet. :p
Quite possible. I'll recheck the Green data sheets on WD's site, and look for the proviso at the bottom. ;)

[yezza]

Just picked these drives up today due to one of my two RAID0 500 GB WD's failing. I've restored to the 1TB's via Time Machine and everything appears to be running great :)

Results 173.02
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 6144 MB
Model MacPro3,1
Drive Type Macintosh HD
Disk Test 173.02
Sequential 209.51
Uncached Write 474.88 291.57 MB/sec [4K blocks]
Uncached Write 481.62 272.50 MB/sec [256K blocks]
Uncached Read 79.24 23.19 MB/sec [4K blocks]
Uncached Read 436.79 219.53 MB/sec [256K blocks]
Random 147.36
Uncached Write 59.29 6.28 MB/sec [4K blocks]
Uncached Write 846.92 271.13 MB/sec [256K blocks]
Uncached Read 191.83 1.36 MB/sec [4K blocks]
Uncached Read 257.43 47.77 MB/sec [256K blocks]


And just for fun, ran this on my mid 2007 Macbook:

Results 40.86
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 1024 MB
Model MacBook2,1
Drive Type ST9120817AS
Disk Test 40.86
Sequential 82.68
Uncached Write 87.29 53.59 MB/sec [4K blocks]
Uncached Write 86.64 49.02 MB/sec [256K blocks]
Uncached Read 65.57 19.19 MB/sec [4K blocks]
Uncached Read 98.72 49.61 MB/sec [256K blocks]
Random 27.13
Uncached Write 8.99 0.95 MB/sec [4K blocks]
Uncached Write 79.99 25.61 MB/sec [256K blocks]
Uncached Read 69.92 0.50 MB/sec [4K blocks]
Uncached Read 107.29 19.91 MB/sec [256K blocks]

[Dave H]

The G4 in my sig. The drive is a 7200RPM PATA with 8MB cache on an ATA-133 card.

Results 48.75
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 1536 MB
Model PowerMac3,4
Drive Type ATA Maxtor 6L120P0
Disk Test 76.79
Sequential 88.21
Uncached Write 82.14 50.43 MB/sec [4K blocks]
Uncached Write 82.68 46.78 MB/sec [256K blocks]
Uncached Read 90.00 26.34 MB/sec [4K blocks]
Uncached Read 100.32 50.42 MB/sec [256K blocks]
Random 67.99
Uncached Write 37.58 3.98 MB/sec [4K blocks]
Uncached Write 88.84 28.44 MB/sec [256K blocks]
Uncached Read 83.89 0.59 MB/sec [4K blocks]
Uncached Read 110.53 20.51 MB/sec [256K blocks]

[Boneoh]

This is for 3 x velociraptor in Raid 0, stock controller.

Results 227.70
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 16384 MB
Model MacPro4,1
Drive Type User
Disk Test 227.70
Sequential 269.35
Uncached Write 533.87 327.79 MB/sec [4K blocks]
Uncached Write 508.08 287.47 MB/sec [256K blocks]
Uncached Read 107.00 31.31 MB/sec [4K blocks]
Uncached Read 601.15 302.14 MB/sec [256K blocks]
Random 197.20
Uncached Write 111.15 11.77 MB/sec [4K blocks]
Uncached Write 510.74 163.51 MB/sec [256K blocks]
Uncached Read 194.28 1.38 MB/sec [4K blocks]
Uncached Read 239.11 44.37 MB/sec [256K blocks]

[MegaMillions]

The first one is my Hitachi 750GB internal storage drive. The second one is the 500GB Maxtor startup disk. Why does the Hitachi have such a problem with random uncached write with 4K blocks? I know it's normal to be terrible at 4K uncached read, but this Hitachi drive has an issue with the random write.

http://img375.imageshack.us/img375/9309/picture2u.png (http://img375.imageshack.us/i/picture2u.png/) http://img375.imageshack.us/img375/picture2u.png/1/w596.png (http://g.imageshack.us/img375/picture2u.png/1/)

http://img43.imageshack.us/img43/4673/picture3mwv.png (http://img43.imageshack.us/i/picture3mwv.png/) http://img43.imageshack.us/img43/picture3mwv.png/1/w596.png (http://g.imageshack.us/img43/picture3mwv.png/1/)

[jamesryanbell]

WD Scorpio Blue 500GB

Disk Test 61.18
Sequential 99.59
Uncached Write 108.58 66.66 MB/sec [4K blocks]
Uncached Write 75.36 42.64 MB/sec [256K blocks]
Uncached Read 96.56 28.26 MB/sec [4K blocks]
Uncached Read 136.42 68.56 MB/sec [256K blocks]
Random 44.15
Uncached Write 16.36 1.73 MB/sec [4K blocks]
Uncached Write 131.88 42.22 MB/sec [256K blocks]
Uncached Read 71.45 0.51 MB/sec [4K blocks]
Uncached Read 126.45 23.46 MB/sec [256K blocks]

[joaoferro37]

STARDOM DECKRAID DR4
Mac Pro 2.8 8 core, 8G ram.
Via FireWire 800, RAID 5, 60MB/s
Via eSATA. 246MB/s

http://www.barefeats.com/hard122.html

STARDOM SOHO Tank ST 8 with High Point 4322 mini SAS, 736MB/s, RAID 5. Got the testing unit from RAIDON-USA
Sequential Read 590MB/S
Sequential Write 437MB/S

[nanofrog]

STARDOM DECKRAID DR4
Mac Pro 2.8 8 core, 8G ram.
Via FireWire 800, RAID 5, 60MB/s
Via eSATA. 246MB/s

http://www.barefeats.com/hard122.html

STARDOM SOHO Tank ST 8 with High Point 4322 mini SAS, 736MB/s, RAID 5. Got the testing unit from RAIDON-USA
Nice. :D

What drives BTW?

[joaoferro37]

Nice. :D

What drives BTW?
The STARDOM DECKRAID testing was using WD 2TB green drive 32MB.
I use hitachi 1TB 16MB with it's built-in RAID 5 via FireWire @ 60MB/S and sometimes I got over 70MB/s
Tried to transfer large file and seems getting faster speed.
Via its SJ20-X8 I got around 246MB/s
Reason I like this unit is it has bult-in RAID 5 for iMac, Macbook Pro.

[VirtualRain]

The STARDOM DECKRAID testing was using WD 2TB green drive 32MB.
I use hitachi 1TB 16MB with it's built-in RAID 5 via FireWire @ 60MB/S and sometimes I got over 70MB/s
Tried to transfer large file and seems getting faster speed.
Via its SJ20-X8 I got around 246MB/s
Reason I like this unit is it has bult-in RAID 5 for iMac, Macbook Pro.

I'm sorry but all this is a bit difficult to interpret... :confused:

How many drives of each kind were used in each of the Stardom tests?

4x 2TB WD Green in the Deckraid?
8x 1TB Hitachi in the Tank?

[joaoferro37]

I'm sorry but all this is a bit difficult to interpret... :confused:

How many drives of each kind were used in each of the Stardom tests?

4x 2TB WD Green in the Deckraid?
8x 1TB Hitachi in the Tank?

Yes, 4 2TB WD green on the stardom deckraid dr4.
8X hitachi 1TB on the stardom decktank dt-8

[Fiveos22]

Whatever is in my imac is getting ~150s,

I am more interested in something tangential: My external HD is getting ready to die (the housing and FW controller as opposed to the actual disk) do you think I should should get an external enclosure for the SATA drive inside or should I just get a new external HD?

Also if I keep the disk and go with the external, should I get a raid enclosure (FW 800) or should I stick with FW400 and start daisy chaining drives. Its been a while since I've had to futz around with my data storage setup (non-mission critical stuff, just the normal backup, and occasional game)

[joaoferro37]

Whatever is in my imac is getting ~150s,

I am more interested in something tangential: My external HD is getting ready to die (the housing and FW controller as opposed to the actual disk) do you think I should should get an external enclosure for the SATA drive inside or should I just get a new external HD?

Also if I keep the disk and go with the external, should I get a raid enclosure (FW 800) or should I stick with FW400 and start daisy chaining drives. Its been a while since I've had to futz around with my data storage setup (non-mission critical stuff, just the normal backup, and occasional game)

The reason I use the stardom deckraid dr4 is it has built-in RAID 5 with quad interface. Perfect for iMac and any Mac laptops.
If you move to Mac Pro, you can still utilize the deckraid dr4 since it has eSATA. The speed thru eSATA is much faster than firewire 800 but it is not an option for iMac.
Besides, the FireWire RAID 1 solution, will take up 50% of the total capacity VS RAID 5 will use 1 drive for parity. Cost per Giga Byte and performance-wise, RAID 5 is a better choice.

[Chad H]

Here is mine with a Intel 80GB Gen2 X-25

[hyram]

STARDOM DECKRAID DR4
Mac Pro 2.8 8 core, 8G ram.
Via FireWire 800, RAID 5, 60MB/s
Via eSATA. 246MB/s
STARDOM SOHO Tank ST 8 with High Point 4322 mini SAS, 736MB/s, RAID 5. Got the testing unit from RAIDON-USA
Sequential Read 590MB/S
Sequential Write 437MB/S

Interesrting results. I'm running a 4322 as well but with the WD 1 TB Blacks. (See posts #52 and #68) You're getting a little more throughput but your xbench numbers are lower across the board. What's you RAID setup???? Block size, stripe size, write and read chache policy, etc??? Also have you run some form of area scan to see what performance you get as the disk fills up???

hyram

[joaoferro37]

RAID 5 stripe size 128K, block size 4K.
Write thru.
Using AJA system, I got much higher speed but the Xbench did not.
Regardless of the benchmark testing, I can capture and edit uncompressed HD footage without any problem.

[joaoferro37]

One question to NanoFrog.
I've been testing the Seagate 1TB, 1.5TB drives using Stardom iTank quad interface enclosure. Under disk utility, it shows only 37G but when I swap out to Hitachi 1TB, it showed correct capacity?
I read some forum saying the seagate drives are problematic?
Can I solve the problem by firmware update or should I never use seagte?

[nanofrog]

One question to NanoFrog.
I've been testing the Seagate 1TB, 1.5TB drives using Stardom iTank quad interface enclosure. Under disk utility, it shows only 37G but when I swap out to Hitachi 1TB, it showed correct capacity?
I read some forum saying the seagate drives are problematic?
Can I solve the problem by firmware update or should I never use seagte?
I'm assuming you're using the consumer models (7200.11 series), and Yes, they've had some problems. As I don't know the age, there's a good chance it is the firmware.

Assuming this is the case, you'd need to do a couple of things.
1. This is a bit of an assumption, but lets assume Seagate doesn't have a DOS image as part of the update untility, or that it works in OS X. So locate a DOS boot disk in order to update the firmware, or create one if you can. (This can be difficult on OS X only systems, as windows is usually needed to create the boot disk). A USB floppy can be used (no windows if you have one), and it makes life so much easier). This is just to get a boot disk made.

If you already have a DOS boot disk, you may not realize just how much ahead of the game you truly are. :eek: :p

2. Contact Seagate, and have them send you a link to the newest firmware revision.

Once you have both of these, you can flash the drive, and it should hopefully solve the problem.

Try this, and see what happens. ;)

[joaoferro37]

I'm assuming you're using the consumer models (7200.11 series), and Yes, they've had some problems. As I don't know the age, there's a good chance it is the firmware.

Assuming this is the case, you'd need to do a couple of things.
1. This is a bit of an assumption, but lets assume Seagate doesn't have a DOS image as part of the update untility, or that it works in OS X. So locate a DOS boot disk in order to update the firmware, or create one if you can. (This can be difficult on OS X only systems, as windows is usually needed to create the boot disk). A USB floppy can be used (no windows if you have one), and it makes life so much easier). This is just to get a boot disk made.

If you already have a DOS boot disk, you may not realize just how much ahead of the game you truly are. :eek: :p

2. Contact Seagate, and have them send you a link to the newest firmware revision.

Once you have both of these, you can flash the drive, and it should hopefully solve the problem.

Try this, and see what happens. ;)

Thank you!! AGAIN!!!! you are the man..
Besides, regarding to the Stardom Deckraid Dr4 benchmark test. Do you think this product is good for iMac user running Final Cut on it?

[nanofrog]

Thank you!! AGAIN!!!! you are the man..
Besides, regarding to the Stardom Deckraid Dr4 benchmark test. Do you think this product is good for iMac user running Final Cut on it?
:cool: No problem. :)

As per the Stardom Deckraid Dr4, I'm under the impression you want graphics/video editing work to reside on a system that can do 250MB/s, so it's close (as in bare min.). Personally, I prefer to plan for worst case, which means you'd want to keep the capacity at 50% or less, as exceeding it will slow you down. Assuming using a solution with additional drives isn't an option (existing hardware has this affect :p), you need to plan for used capacity = 50% or less of the array capacity. So drive selection would be key IMO. ;) :D

[joaoferro37]

I am a consultant and most of my clients are graphic designers and film maker wanna be. Some of them are using iMac and they want to have RAID solution to protect their photos and footage. Burning DVDs is just painful and RAID 1 solution will lose 50% of the capacity and it is slow.

I tried to bring them CalDigit FireWire VR but this company is not as reliable as I thought. They phase out product too frequent and I got no support after that.

I am still testing the deckraid DR4 and so far I am happy with it. RAIDON sent me a loaner unit for testing the beauty of this item is it provides the missing capability of RAD 5 features for Mac mini and iMac.

WAIT!! why do people want to buy a iMac or Mac mini for graphic design!!!

[nanofrog]

I am a consultant and most of my clients are graphic designers and film maker wanna be. Some of them are using iMac and they want to have RAID solution to protect their photos and footage. Burning DVDs is just painful and RAID 1 solution will lose 50% of the capacity and it is slow.

I tried to bring them CalDigit FireWire VR but this company is not as reliable as I thought. They phase out product too frequent and I got no support after that.

I am still testing the deckraid DR4 and so far I am happy with it. RAIDON sent me a loaner unit for testing the beauty of this item is it provides the missing capability of RAD 5 features for Mac mini and iMac.

WAIT!! why do people want to buy a iMac or Mac mini for graphic design!!!
I'm no fan of CalDigit's gear either due to a disaster with their RAID card. :rolleyes: :mad:

As for using a Mini or iMac, it does seem like a major contradiction. ;)
Perhaps limited budget (especially students), and they want to use Mac based software?
Hobby?

[hyram]

One question to NanoFrog.
I've been testing the Seagate 1TB, 1.5TB drives using Stardom iTank quad interface enclosure. Under disk utility, it shows only 37G but when I swap out to Hitachi 1TB, it showed correct capacity?
I read some forum saying the seagate drives are problematic?
Can I solve the problem by firmware update or should I never use seagte?

I'd steer clear of the Seagate 7200.11 drives for RAID. Check out their forum; it's littered with the dead bodies of folks trying to get them to work... including mine :) The drives have a habit of dropping out of the array. There has been no firmware update that will fix the problem either. The ES2 drives on the otherhand seem to be rock solid if you're willing to pay for them.

hyram

[nanofrog]

I'd steer clear of the Seagate 7200.11 drives for RAID. Check out their forum; it's littered with the dead bodies of folks trying to get them to work... including mine :) The drives have a habit of dropping out of the array. There has been no firmware update that will fix the problem either. The ES2 drives on the otherhand seem to be rock solid if you're willing to pay for them.

hyram
As a general rule, enterprise RAID drives should be used, as they're built for that purpose. ;) They do tend to be far more stable, as the recovery settings in the drive firmware are different (7,0 rather than 0,0 used in consumer models). WD is the only make I've ever found the utility to adjust these values.

As it happens, the firmware issues with the Seagate's also affected the ES.2's as well (boot of death). :( Though they are far more stable than their consumer counterparts.

[joaoferro37]

As a general rule, enterprise RAID drives should be used, as they're built for that purpose. ;) They do tend to be far more stable, as the recovery settings in the drive firmware are different (7,0 rather than 0,0 used in consumer models). WD is the only make I've ever found the utility to adjust these values.

As it happens, the firmware issues with the Seagate's also affected the ES.2's as well (boot of death). :( Though they are far more stable than their consumer counterparts.
I never have any issue with Hitachi and I always suggest my clients to get Hitachi. Correct me if I am wrong.

[nanofrog]

I never have any issue with Hitachi and I always suggest my clients to get Hitachi. Correct me if I am wrong.
If you've never needed to deal with them for firmware, that's wonderful. :) Unfortunately, I have, and was told bluntly they don't support firmware issues, and had to deal with the vendor. One specific occasion, happened to be CalDigit's HDElement enclosure, which was stuffed with 1TB consumer models (7K1000 IIRC). They couldn't deal with the drive firmware either, and didn't seem to have a clue how to go about working it out with Hitachi. (It was a typical consumer drive, and didn't have the proper recovery timings for RAID). Now I do place the bulk of the situation on CalDigit for poor drive selection, but Hitachi could have considered the end user's issues better, and used a better degree of tact.

I've run into similar situations with the Ultrastar line (enterprise models) needing the firmware revision to work with a specific RAID card (usually a SAS model). I was always directed to contact the controller manufacturer for the correct version. Why should they have it, as they didn't make the drive?

It would have been one thing had it been like CalDigit, where I was trying to use a consumer drive for enterprise use, but I had the correct line. So why did they feel the need to be such a PITA about giving me the right firmware revision?

This pissed me off something fierce, and I now refuse to use their drives, as firmware does come up from time to time (needs flashed, usually for a newer version that was verified as PASS by the card manufacturer). :mad: :mad:

[joaoferro37]

I cannot believe that CalDigit's profit margin is very high and they provide poor drives to end users in order to Squeeze out more profit?
For example, a 4 drive enclosure is around 300 dollars range, ex: Stardom, Proavio, OWC, EZquest. A miniSAS RAID card is arond 500 dollars range. CalDigit is selling a 4TB with card for $2200.
If I am not mistaken, 300+500+500(for the drive), their profit margin is around $900.
Can we have better service and product and not feel got ripped off?

[nanofrog]

I cannot believe that CalDigit's profit margin is very high and they provide poor drives to end users in order to Squeeze out more profit?
For example, a 4 drive enclosure is around 300 dollars range, ex: Stardom, Proavio, OWC, EZquest. A miniSAS RAID card is arond 500 dollars range. CalDigit is selling a 4TB with card for $2200.
If I am not mistaken, 300+500+500(for the drive), their profit margin is around $900.
Can we have better service and product and not feel got ripped off?
I'd think so, but apparently not from CalDigit. :eek: They're apparently way too greedy. :D :p

BTW, their costs are lower than that, as they get products at quantity pricing (they don't make any of it). So the margins are even higher. :eek: ;)

[joaoferro37]

Computer: Mac Pro 2.8 8-core, 4G Ram
OS: 10.5.7
HDD: Hitachi 1TB, 16Mb cache. model: OA38016
RAID Card 2: High Point 4322 miniSAS @4lane slot
RAID level:0
Seq Write 548MB/s
Seq Read 605MB/s
Random Write 489MB/s
Random Read 616MB/s
RAID Level: 5
Seq Write 353MB/s
Seq Read 6634MB/s
Random Write 251MB/s
Random Read 95MB/s
186725
186726

[nanofrog]

Computer: Mac Pro 2.8 8-core, 4G Ram
OS: 10.5.7
HDD: Hitachi 1TB, 16Mb cache. model: OA38016
RAID Card 2: High Point 4322 miniSAS @4lane slot
RAID level:0
Seq Write 548MB/s
Seq Read 605MB/s
Random Write 489MB/s
Random Read 616MB/s
RAID Level: 5
Seq Write 353MB/s
Seq Read 6634MB/s
Random Write 251MB/s
Random Read 95MB/s

Enable the disk cache setting on AJA, and re-run the tests. You might like the results. ;) :D

[joaoferro37]

Let me try, I am getting Areca 1680 card for testing as well as High Point 2314 and High Point MAc SATA card. Will post it in 30 minutes.

[ji3yjo42ul3]

Computer: Mac Pro 2.8 8-core, 4G Ram
OS: 10.5.7
HDD: Hitachi 1TB, 16Mb cache. model: OA38016
RAID Card 2: High Point 4322 miniSAS @4lane slot
RAID level:0
Seq Write 548MB/s
Seq Read 605MB/s
Random Write 489MB/s
Random Read 616MB/s
RAID Level: 5
Seq Write 353MB/s
Seq Read 6634MB/s
Random Write 251MB/s
Random Read 95MB/s
186725
186726

It sounds really interesting to me. I am taking wedding video, I am looking for storage product with RAID function. But I have no idea about how it really works and how to set it up. From your testing result, Stardom DR4-WBS2 seems it's fast enough and safe to my current work loading. :rolleyes:

I want to buy it, is website : www.stardom.com.....but I can't find it there. :confused: It's an American company, right? Where can I buy from? I am located in South of California.

Ji3

[ji3yjo42ul3]

I cannot believe that CalDigit's profit margin is very high and they provide poor drives to end users in order to Squeeze out more profit?
For example, a 4 drive enclosure is around 300 dollars range, ex: Stardom, Proavio, OWC, EZquest. A miniSAS RAID card is arond 500 dollars range. CalDigit is selling a 4TB with card for $2200.
If I am not mistaken, 300+500+500(for the drive), their profit margin is around $900.
Can we have better service and product and not feel got ripped off?


Do you think I can buy a nice 4.5TB(1.5TB per drive) RAID5 subsystem under $1200?

[joaoferro37]

It sounds really interesting to me. I am taking wedding video, I am looking for storage product with RAID function. But I have no idea about how it really works and how to set it up. From your testing result, Stardom DR4-WBS2 seems it's fast enough and safe to my current work loading. :rolleyes:

I want to buy it, is website : www.stardom.com.....but I can't find it there. :confused: It's an American company, right? Where can I buy from? I am located in South of California.

Ji3

It depends on what footage you are working on. I assuming for wedding projects, mostly, it will be HDV or DVCPro HD. If that is the case, stardom DR4 is fast enough for sure and you cannot beat the built-in RAID 5 feature.
I believe the website is http://www.raidon-usa.com/stardom/index.htm
You can find their product from OWC, www.macsales.com

[joaoferro37]

Enable the disk cache setting on AJA, and re-run the tests. You might like the results. ;) :D

Using High Point SATA card + stardom deckraid dr4 via esata (4 hitachi 1TB 16MB)
The number cannot be right!!!!

[nanofrog]

Let me try, I am getting Areca 1680 card for testing as well as High Point 2314 and High Point MAc SATA card. Will post it in 30 minutes.
The ARC-1680 series are really nice cards, and have features the Highpoint doesn't. It will outperform the Highpoint as well, especially if you have a model you can upgrade the cache (Areca designed the 4xxx series BTW).

Do you think I can buy a nice 4.5TB(1.5TB per drive) RAID5 subsystem under $1200?
It's possible, but it will depend on exactly what you're after.

How many drives for future expansion?
What kind of speed?
Will you need to migrate to a different array type?
What OS support do you need (drivers), and do you plan to boot from it (if so, which OS's)?

The answers will make a difference. Also, you should seriously consider enterprise drives, as consumer models have a habit of being unstable on hardware solutions, especially SAS based models.

Using High Point SATA card + stardom deckraid dr4 via esata (4 hitachi 1TB 16MB)
The number cannot be right!!!!
That's the influence of the cache (when the test files all fit on the cache, you get insane speeds). Try increasing the file size to larger than the cache on the card. Then you should get something more realistic.

[joaoferro37]

Enable the disk cache setting on AJA, and re-run the tests. You might like the results. ;) :D

WEIRD!!!
2G file size, system 4G
With system cache off, I got 690MB write and 721MB read
With system cache on, I got 495MB write and 1355MB read

[nanofrog]

WEIRD!!!
2G file size, system 4G
With system cache off, I got 690MB write and 721MB read
With system cache on, I got 495MB write and 1355MB read
I've seen this before, and to some extent, is what happens. The adjusting the settings might use a bit of tweaking to suit your needs.

What is the cache setting on the card?
Other settings as well, such as the stripe size, disk operation, NCQ/TCQ,...?

[joaoferro37]

I've seen this before, and to some extent, is what happens. The adjusting the settings might use a bit of tweaking to suit your needs.

What is the cache setting on the card?
Other settings as well, such as the stripe size, disk operation, NCQ/TCQ,...?

I know I cannot mess around with Mr. frog when the topic is about speed and drives.
The strip size is at 128K block size, NCQ enable, Read ahead cache on, write cache disable. RAID card memory, 256.<<<If I can change it to 512 or 1G, it will make huge difference. RAID card 800MHz IOP348 <<< Areca has 1.2GHz version which I have not get a chance to test it yet.

I like the Areca card over High Point because the tech support from High Point is a nightmare. Even for the testing cards that I got from them were not updated and their firmware update requires a pure dos boot up environment.
I believe many of you had the same experience that a pure PC mentality company wants to do Mac business and never study what we really needs and I truly believe they don't even have a Mac in their testing lab.

[joaoferro37]

Do you think I can buy a nice 4.5TB(1.5TB per drive) RAID5 subsystem under $1200?

To get a 4.5TB with 1.5TB.. That means you have to get Seagate.. Ask Mr. Frog see if he suggest using seagate, he will say NO!!!

MY suggestion is getting a Areca 1221X card with Stardom Decktank DT-8 $695
http://eshop.macsales.com/item/Raidon/ST8S2P/
and get the 2314 $195
http://eshop.macsales.com/item/Highpoint%20Technologies/RR2314/
1TB $84
http://eshop.macsales.com/item/Hitachi/0A38016/
Total $1562
A 8TB with RAID 5 for $1562, i cannot complain.

There are alternative solutions but I trust OWC.

[nanofrog]

I know I cannot mess around with Mr. frog when the topic is about speed and drives.
The strip size is at 128K block size, NCQ enable, Read ahead cache on, write cache disable. RAID card memory, 256.<<<If I can change it to 512 or 1G, it will make huge difference. RAID card 800MHz IOP348 <<< Areca has 1.2GHz version which I have not get a chance to test it yet.
1. You could try setting the stripe to 256K (it doesn't seem to be worth going larger IMO)
2. Disable NCQ
3. Enable Write Cache, which can be done in the browser window, but it also needs to be set as write through, which is set in the card's firmware

Test it out and see what happens. You can play with the NCQ setting, as sometimes it can help a little, but be careful, as it can also cause instability, particularly on the SAS models. (You need to understand how the card behaves anyway, including simulating various failures, like pull the power cord (assuming it's attached to a UPS), and pull a drive and see how it reacts (degraded mode, rebuilds,...). If you're unsure of the details of your useage pattern (types and frequency of file transfers), nows the time to find out. :p

I like the Areca card over High Point because the tech support from High Point is a nightmare. Even for the testing cards that I got from them were not updated and their firmware update requires a pure dos boot up environment.
I believe many of you had the same experience that a pure PC mentality company wants to do Mac business and never study what we really needs and I truly believe they don't even have a Mac in their testing lab.
Highpoint also has the issue you can only have EFI or BIOS in it, not both. So OS boot capability is affected by this. Unfortunately, most firmware updates are still done on the PC side (DOS), as it's been around, and it does work. But they assume user's have access to a DOS boot disk, and this can be a problem for Mac users who don't run windows or have access to a windows machine/friend who's willing to make one for them.

To get a 4.5TB with 1.5TB.. That means you have to get Seagate.. Ask Mr. Frog see if he suggest using seagate, he will say NO!!!

MY suggestion is getting a Areca 1221X card with Stardom Decktank DT-8 $695
http://eshop.macsales.com/item/Raidon/ST8S2P/
and get the 2314 $195
http://eshop.macsales.com/item/Highpoint%20Technologies/RR2314/
1TB $84
http://eshop.macsales.com/item/Hitachi/0A38016/
Total $1562
A 8TB with RAID 5 for $1562, i cannot complain.

There are alternative solutions but I trust OWC.
1. I don't like what happened over the 7200.11 models, particularly the 1.5TB model. :eek: It was a disaster. :rolleyes: So I don't trust those. :( Others may not be an issue though (Samsung), but research is warranted.

The ARC-1221X may work, so long as you don't need to boot from it (driver support only for Mac). If a PM's limitations are acceptable, that's not a bad setup, though there are other enclosures that are really good, but a bit less money. An Enhance E8 PM (http://www.directron.com/enhanceboxe8pm.html) unit should be ~$100 - 150USD less. Others as well, that work well, and actually look decent too.

Otherwise, another card and enclosre (port connections) would be needed (greater throughput potential).

[hyram]

Even for the testing cards that I got from them were not updated and their firmware update requires a pure dos boot up environment.

Ok... please explain. I've had a HPT4322 since Dec 08. I've updated the firmware on it twice and did not have to use DOS to do so. This can be done using the web interface only.

hyram

[lemonade-maker]

To get a 4.5TB with 1.5TB.. That means you have to get Seagate.. Ask Mr. Frog see if he suggest using seagate, he will say NO!!!

Its a firmware issue, if you have the latest, no prob. My 6 seagate 1.5s are just fine. So are many others. Not one problem which is far from what I can say about all the wd junk I've had. They are solid drives and i even picked up a couple extra. They are cheap and perform well in software raid0.

[gugucom]

Results 403.60
System Info
Xbench Version 1.3
System Version 10.5.8 (9L30)
Physical RAM 12288 MB
Model MacPro1,1
Drive Type OS X RAID
Disk Test 403.60
Sequential 308.52
Uncached Write 394.61 242.29 MB/sec [4K blocks]
Uncached Write 610.54 345.45 MB/sec [256K blocks]
Uncached Read 134.84 39.46 MB/sec [4K blocks]
Uncached Read 726.21 364.99 MB/sec [256K blocks]
Random 583.36
Uncached Write 238.92 25.29 MB/sec [4K blocks]
Uncached Write 712.01 227.94 MB/sec [256K blocks]
Uncached Read 2898.2220.54 MB/sec [4K blocks]
Uncached Read 1084.74201.28 MB/sec [256K blocks]

[gugucom]

Here's my RAMDisk:
Results 906.19
System Info
Xbench Version 1.3
System Version 10.5.7 (9J61)
Physical RAM 12288 MB
Model MacPro1,1
Drive Type Apple read/write RAM-Drive

Disk Test 906.19
Sequential 553.47
Uncached Write 415.27 254.97 MB/sec [4K blocks]
Uncached Write 1242.03 702.74 MB/sec [256K blocks]
Uncached Read 282.08 82.55 MB/sec [4K blocks]
Uncached Read 2132.99 1072.02 MB/sec [256K blocks]
Random 2498.42
Uncached Write 1195.78 126.59 MB/sec [4K blocks]
Uncached Write 2303.16 737.33 MB/sec [256K blocks]
Uncached Read 8109.22 57.46 MB/sec [4K blocks]
Uncached Read 4825.48 895.40 MB/sec [256K blocks]

Hi Tess, could you explain how you set up a RAM disk in OS X?

[joaoferro37]

[QUOTE=hyram;8236270]Ok... please explain. I've had a HPT4322 since Dec 08. I've updated the firmware on it twice and did not have to use DOS to do so. This can be done using the web interface only.

hyram[/QUOTE
I happen to test high point 2314, 3522 4322 cards for a post production company.
The problem we have on the 3522 was the firmware was not compatible with 10.5.7 at the time which means the card won't show up on the system profiler.
I contact high point and they told me I had to find a PC boot up from pure windows to update.
The first 4322 card I test was direct from them and I had to do the same thing.
I remove the driver and plist restart, reinstall and tried so many times, the system did not see it. It cannot communicate with the card changing the port from2704 to 2714, etc. I had to do the same thing update it from a pure windows dos. After that, the problem went away.

[hyram]

After that, the problem went away.

Perhaps you had a pre-production version, the card only came out last Nov. Honesty, I've had zero problems with my 4322. Setup was easy, and updating was very simple. I'll agree that the web interface is a little cryptic at times and the documentation is horrible, but that seems to be a common thread with all the far east companies. If you can you should give it another look, I don't think you'll be disappointed with the card itself.

Tech service? Well that's another story, I've not had reason to call/email Highpoint but I've heard things; but then I've heard the same about Areca, Atto, and Caldigit as well. And one bad call will taint you for life.

[nanofrog]

Perhaps you had a pre-production version, the card only came out last Nov. Honesty, I've had zero problems with my 4322. Setup was easy, and updating was very simple. I'll agree that the web interface is a little cryptic at times and the documentation is horrible, but that seems to be a common thread with all the far east companies. If you can you should give it another look, I don't think you'll be disappointed with the card itself.

Tech service? Well that's another story, I've not had reason to call/email Highpoint but I've heard things; but then I've heard the same about Areca, Atto, and Caldigit as well. And one bad call will taint you for life.
I've had decent luck with Areca. Though it's been strictly emails, as a phone call isn't possible. I don't speak the language, and the emails are cryptic as is. I can usually figure it out, but it does take a little longer due to the language barrier. Atto's Tech Support is US based. At least those I've dealt with. Highpoint's are as well (both are in NY, IIRC). Keep in mind though, Areca and Atto actually do their own design and manufacturing.

CalDigit's horrible. They and Highpoint just have other companies (ODM) do the work. Accusys does the work for CalDigit, and Highpoint sources multiple companies. Areca's one of them (43xx series).

[joaoferro37]

Mr. Nano. I am trying to work on this crazy idea of building the fastest Mac.
Thinking putting 4 Stardom Pro Drive set to RAID 0 with 2 SSD, total of 8
then hook up to a internal RAID card or using OS stripe.
I know the stardom pro drive can set up raid 0 and SSD seems to be the fatest option.
The reason to build this machine is to increase the render speed for final cut pro, motion, premiere, Maya using internal SSD.

[Tesselator]

To get a 4.5TB with 1.5TB.. That means you have to get Seagate.. Ask Mr. Frog see if he suggest using seagate, he will say NO!!!


Wasn't Samsung the first out with 1.5 TB drives? I dunno it they were first or not (I thought they were) but they were surely the first out with 500GB platter density. I'm using them in a 3-drive RAID0 with great success. Nano tried to buy these units as well but at the time they were higher than he wanted to spend. They're now about $110 ~ $120 each in the USA so well within the $1200 joaoferro37 mentioned.


Hi Tess, could you explain how you set up a RAM disk in OS X?

Sure. I just downloaded and installed this: http://www.macupdate.com/info.php/id/16518

It works pretty good and even backs itself up on a shutdown and restores on boot. :cool:

[nanofrog]

Mr. Nano. I am trying to work on this crazy idea of building the fastest Mac.
Thinking putting 4 Stardom Pro Drive set to RAID 0 with 2 SSD, total of 8
then hook up to a internal RAID card or using OS stripe.
I know the stardom pro drive can set up raid 0 and SSD seems to be the fatest option.
The reason to build this machine is to increase the render speed for final cut pro, motion, premiere, Maya using internal SSD.
I'm not quite following how you're wanting to set this up.

You mean to use 8x SSD's in a stripe? Or something else?
And how much throughput are you looking for? Or is this a "just to see" type of experiment?

Wasn't Samsung the first out with 1.5 TB drives? I dunno it they were first or not (I thought they were) but they were surely the first out with 500GB platter density. I'm using them in a 3-drive RAID0 with great success. Nano tried to buy these units as well but at the time they were higher than he wanted to spend. They're now about $110 ~ $120 each in the USA so well within the $1200 joaoferro37 mentioned.
I thought it was Seagate, as the platter density was lower (7200.11 series platters). Followed by Samsung with the 500GB platters. Either way, I'd still give the Samsung a try, as the fiasco with the 7200.11's have left me nervous about Seagate in general (as the issues affected the ES.2 enterprise line as well).

[joaoferro37]

I am trying to use 8 or even 10 (using optical bay) to make a OS stripe drive for extremely fast scratch disk.

I can use the stardom pro drive for my mac pro.
This product takes 2 SSD or 2.5" sata drive in one mac pro drive bay.
Has built-in RAID 0 or RAID 1.

I am thinking if this can reach over 1GB per second!!!
Yes this will solve some problems for my clients who are in need of high speed.

[jethrodesign]

Hi, these are our results on our server. It's using an older Highpoint RR1820a setup with 3 Seagate 200GB drives (yeah, it's pretty old now) running a RAID 5.

Results 27.98
System Info
Xbench Version 1.3
System Version 10.4.11 (8S169)
Physical RAM 1024 MB
Model PowerMac3,1
Processor PowerPC G4 @ 0 MHz
Version 7455 (Apollo) v3.2
L1 Cache 32K (instruction), 32K (data)
L2 Cache 256K
L3 Cache 2048K
Bus Frequency 100 MHz
Video Card ATY,RV250
Drive Type RR182x RAID 5 Array

Disk Test 27.98
Sequential 84.41
Uncached Write 70.04 43.00 MB/sec [4K blocks]
Uncached Write 83.94 47.49 MB/sec [256K blocks]
Uncached Read 59.16 17.31 MB/sec [4K blocks]
Uncached Read 232.99 117.10 MB/sec [256K blocks]

Random 16.77
Uncached Write 5.06 0.54 MB/sec [4K blocks]
Uncached Write 43.48 13.92 MB/sec [256K blocks]
Uncached Read 95.89 0.68 MB/sec [4K blocks]
Uncached Read 134.97 25.04 MB/sec [256K blocks]



A) The drives are pretty full, but is this still a bit low for a RAID 5?

We're migrating our server to a G5 1.6Ghz machine. We'll probably still use the RR1820 as this machine has only standard PCI slots. Probably get WD RE3 drives.

B) Any better setup we should consider to get some better performance while still having security/redundancy as our #1 priority?

Considering the following:
- Simple RAID 1 using Apple software RAID
- RAID 10 with 4 drives using Highpoint controller
- RAID 5 with 3 or 4 drives using Highpoint controller (similar to what we have).

This server is almost entirely used for just serving files to client machines over gigabit ethernet. Never worked on directly.

[nanofrog]

I am trying to use 8 or even 10 (using optical bay) to make a OS stripe drive for extremely fast scratch disk.

I can use the stardom pro drive for my mac pro.
This product takes 2 SSD or 2.5" sata drive in one mac pro drive bay.
Has built-in RAID 0 or RAID 1.

I am thinking if this can reach over 1GB per second!!!
Yes this will solve some problems for my clients who are in need of high speed.
Unfortunately, that's not going to work. :( The Pro Drive isn't going to give you the necessary throughput, as it's 2x drives to a single SATA port. To do this, it uses a PM chip, which at best, usually gives 250MB/s. You'd need 500MB/s.

Stardom's spec of up to 300MB/s, is very misleading, as that's the single drive operation. (And SATA 3.0gB/S = 375MB/s BTW; divide 3000 by 8).

You can still do it though (still internal), but you'd need to take a different route.

Use a 4x 2.5" backplane unit (http://www.newegg.com/Product/Product.aspx?Item=N82E16816215098) (fits a single 5.25" bay) in the empty optical bay. Then use 2.5" to 3.5" adapters for the HDD bays (1 - 4). This keeps you at 1 SATA port per drive, which is what you need to get the throughput you want out of SSD's.

For the 2.5" to 3.5" adapters, there's two that will work.

MaxUpgrades SSD/Velociraptor Mount (http://www.maxupgrades.com/istore/index.cfm?fuseaction=Product.display&product_id=180&CFID=2524326&CFTOKEN=3618004) (they have versions that fit '06 - '09 MP's; '09's are different than the others)


There's another by IcyDock (http://www.icydock.com/product/mb882sp-1s-1.html) that will fit as well (fits the existing tray in the system, so works on all models as well, and it's less expensive as well)


This setup still gives you 8x SSD's, all with 1 port each. :) Any more though, and you'd have to pull the optical drive (place it in an external 5.25" enclosure), for access to that internal bay. Then you could add an additional 4, bringing the total to 12. You'd also need a different RAID card, and that would be the ARC-1680ix12 (or larger, as you can still add externally).

[joaoferro37]

I was thinking using Stardom Pro Drive and utilize its build-in RAID 0 then do OS stripe.
That means I can utilize 8 SSD.
It ain't gonna work?

[nanofrog]

I was thinking using Stardom Pro Drive and utilize its build-in RAID 0 then do OS stripe.
That means I can utilize 8 SSD.
It ain't gonna work?
Yes and No. :eek:

That is, the Pro Drive (sold by more than Stardom BTW), will allow you to run 8x drives in the HDD bays on the MP. So there's the Yes. :p

The No, however, is the fact that the RAID0 is achieved via a Port Multiplier chip, and it will throttle the throughput. Let's say it acts as most do, and gives a max throuput of 250MB/s. That's going to be for BOTH drives COMBINED, not each. It may be less. Maybe 1GB/s is possible, if you're lucky. But I do recall somewhere, the Pro Drive is more like 200MB/s, so not that high.

I wouldn't trust it, when there's other options. And better still, the methodology I described is cheaper than the Pro Drives (x4)

$60 + 4x $25 = $160USD Total. (all of it from newegg; icy dock adapter (http://www.newegg.com/Product/Product.aspx?Item=N82E16817994064), 4x 2.5" backplane enclosure (http://www.newegg.com/Product/Product.aspx?Item=N82E16816215098)

So to get the max throughput, EACH SSD needs it's own SATA port (not shared).

[msmth928]

I have 2 500GB 16MB Cache Seagates that both average at around 85

Just a quick note to say I am now running SL and with a clean install put those two drives into Raid stripe. Xbench now gives me (3 goes..)

131
131
136

Not quite double the performance but still pretty neat :)

[justit]

I'm using Esperance DV 2.3.2 that I got from here: http://www.macupdate.com/info.php/id/16518

....

PS is still installed on it after several reboots and I was even able to shrink it's size down closer to PS's needs - which for me is 800MB. :)

As for the naysayers saying RAM Disks such as this aren't fast just look no further than the benchmark. How useful it is depends on you of course.

1151.59 Disk

@Tesselator This Ram Disk is set to 2GB. Are you suggesting to make only an 800MB ram disk? Isn't larger the better? I work with many small (200K) website images.

[zachsilvey]

I scored about 125 as an average of 3 runs on a Raid 5 configuration of 4x1tb hitachi deskstar 7200 rpm using Apple's raid card.