Notes on Performance, Benchmarks and Cache.

raymod2

Dabbler
Joined
Dec 14, 2011
Messages
15
Are you guys sure this "dd" method produces accurate results? I ran the following two commands on my FreeNAS system:

dd if=/dev/zero of=tmp.dat bs=2048k count=50k
dd if=tmp.dat of=/dev/null bs=2048k count=50k

The results were as follows:

107374182400 bytes transferred in 694.660328 secs (154570771 bytes/sec)(write)
107374182400 bytes transferred in 722.927058 secs (148526993 bytes/sec)(read)

My system specs are as follows:

SuperMicro MBD-X9SCM-F-O (C204 motherboard)
Intel Pentium G850 (2.9GHz dual core CPU)
Kingston KVR1333D3E9SK2/8G (2x4G DDR3 1333 ECC RAM)
Western Digital WD20EARX (2Tx4 in a RAIDZ-1 configuration)

These benchmark numbers are low considering each drive is capable of 110 MB/s sustained transfer rates. I would expect up to 330 MB/s since the data is striped across three drives with one drive for parity. The next thing I tried is copying 861GB of large files to the ZFS pool and then scrubbing it. The scrub only took 57 minutes. That corresponds to 270 MB/sec which sounds much more reasonable.
 

matsoh

Cadet
Joined
Oct 27, 2011
Messages
7
dd if=/dev/zero of=tmp.dat bs=2048k count=50k
dd if=tmp.dat of=/dev/null bs=2048k count=50k

Here are the results:
WRITE: 107374182400 bytes transferred in 997.424408 secs (107651449 bytes/sec)
READ: 107374182400 bytes transferred in 883.145245 secs (121581567 bytes/sec)

Those are not very good, probably due to my 4 years old Harddisks. I've planned to change them for 3xHitachi 5k3000 2To

My configuration:
- Supermicro X7SPA-HF-D525 (Atom D525, Hyperthreading: off)
- 2x Corsair DDR3 2 Go PC10600
- 3x Seagate Barracuda 7200.11 500 Go
- Lian-Li PC-Q11
- Antec Truepower New 550w
- Netgear GS108T v2
- FreeNAS 8.0.2 / RaidZ / LACP
 

thag

Cadet
Joined
Dec 31, 2011
Messages
1
Hi All,

First i have some comments on this post.

1) Disks normally used at home have a blocksize of 512k
2) when dealing with /dev/zero (zero's) you will have a caching issue that clouds your performance, zero's have to be committed first to clear the cache for reads.

# Write
dd if=/dev/zero of=/<path of your store>/test.dd bs=512k count=20480 conv=sync
# Read
dd if=/<path of your store>/test.dd of=/dev/zero bs=512k count=20480 conv=sync

This will create a zero'd file of 10Gb using 512k blocksize (mostly default for example: Western Digital Green Disks or other normal disks.)

## Note from dd (1) man page for FreeBSD
sync Pad every input block to the input buffer size. Spaces
are used for pad bytes if a block oriented conversion
value is specified, otherwise NUL bytes are used.
## END note

We use these kind of tests often in our enterprise environment.

Now for my output using the dd read / write examples above.
I must say that i have added a loader to disable prefetch.

# Write
# /mnt/pool/share# dd if=/dev/zero of=tmp.dd bs=512k count=20480 conv=sync
20480+0 records in
20480+0 records out
10737418240 bytes transferred in 74.581595 secs (143968740 bytes/sec)

# Read
# /mnt/pool/share# dd if=tmp.dd of=/dev/zero bs=512k count=20480 conv=sync
20480+0 records in
20480+0 records out
10737418240 bytes transferred in 52.085372 secs (206150360 bytes/sec)

Write: 137.299289703369 megabytes/s (done in 74.6 seconds)
Read: 196.600303649902 megabytes/s (done in 52.1 seconds)

Hardware:
Zocom ID-41
CPU: Intel® Atom™ D525 (dual-core) (1.8 GHz)
Memory: 4GB

Storage:
Sharkoon 5 bay raidbox 4 disks of 1 TB in raid5 mode.

You can get your blocksize using diskinfo

[root@host] # diskinfo /dev/ada0
/dev/ada0 512 3000470863872 5860294656 0 0 5813784 16 63

2nd block is the blocksize of your disk.

My 10 cents.

\richard
 

b1ghen

Contributor
Joined
Oct 19, 2011
Messages
113
1) Disks normally used at home have a blocksize of 512k

/dev/ada0 512 3000470863872 5860294656 0 0 5813784 16 63

2nd block is the blocksize of your disk.


Isn't the block (sector) size 512 bytes (B) in this example? not 512 kilobytes (kB) like you are saying?
And many popular newer consumer disks use the 4kB block size and some 512B emulation but that discussion is too much to bring back in this thread on new years eve :)
 

Gernash

Cadet
Joined
Jan 7, 2012
Messages
5
ZFS Raidz2
Adaptec 6805 8x2Tb 7200rpm Seagate Hdd's
Gigabyte socket 1156 with 8Gb DDR3


dd if=/dev/zero of=tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 314.637988 secs (341262614 bytes/sec)

dd if=tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 152.155110 secs (705688968 bytes/sec)
 

raulfg3

Dabbler
Joined
May 27, 2011
Messages
40
NAS-A=On Board SAT=1TB Disk
NAS-B=LSI SAS3081E-R=2TB disk

freenas:/mnt/NAS-A/Temp# dd if=/dev/zero of=/mnt/NAS-A/Temp/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 538.082417 secs (199549695 bytes/sec)
freenas:/mnt/NAS-A/Temp#

freenas:/mnt/NAS-B/Temp# dd if=/dev/zero of=/mnt/NAS-B/Temp/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 466.315429 secs (230260840 bytes/sec)
freenas:/mnt/NAS-B/Temp#


Lectura:

freenas:/mnt/NAS-B/Temp# dd if=/mnt/NAS-B/Temp/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 260.578697 secs (412060478 bytes/sec)


freenas:/mnt/NAS-B/Temp# dd if=/mnt/NAS-A/Temp/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 428.078355 secs (250828338 bytes/sec)
freenas:/mnt/NAS-B/Temp#
 

Daisuke

Contributor
Joined
Jun 23, 2011
Messages
1,038
FreeNAS 8.0.3 on an Intel Atom D525 1.8GHz processor with 8GB RAM and ZFS RaidZ2 array of 6 disks + 1 64GB SSD cache disk (see complete setup):

Code:
$ dd if=/dev/zero of=/mnt/nas/media/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 602.091800 secs (178335235 bytes/sec)

$ dd if=/mnt/nas/media/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 363.791025 secs (295153467 bytes/sec)


I get better results with hyper-threading enabled, anything else related to configuration has stock settings. There is a myth stating that Atom processors cannot handle more than 4GB of RAM. My system uses on a regular basis 6GB of Wired RAM, which is technically impossible based on Intel specs.

 

McFly

Cadet
Joined
Dec 9, 2011
Messages
5
What is the discussion on compression? I got:

[root@freenas] /mnt/BIGRAID# dd if=/dev/zero of=tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 82.832872 secs (1296275016 bytes/sec)

That seems amazingly impossible based on the other numbers - is that my compression fooling me?

Random I got:
[root@freenas] /mnt/BIGRAID# dd if=/dev/random of=tmp.dd bs=512k count=20480 conv=sync
20480+0 records in
20480+0 records out
10737418240 bytes transferred in 158.649046 secs (67680320 bytes/sec)
 

gian4nas

Patron
Joined
Oct 13, 2011
Messages
389
The result of mine FreeNAS 8.0.4 multimedia AMD Athlon(tm) II Neo N36L Dual-Core Processor with 4G ram and storage of 3x2Tb WD green in RAID_Z1

/mnt/Archivio# dd if=/dev/zero of=tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 642.232426 secs (167188977 bytes/sec)

dd if=tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 552.289443 secs (194416503 bytes/sec)

/mnt/Archivio# dd if=/dev/random of=tmp.dd bs=512k count=20480 conv=sync
20480+0 records in
20480+0 records out
10737418240 bytes transferred in 299.856821 secs (35808484 bytes/sec)
 

Stephen J

Dabbler
Joined
Feb 3, 2012
Messages
49
dd if=/dev/zero of=/mnt/MORODATA/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 619.197167 secs (1743408711 bytes/sec) - 165.37 MB/sec

dd if=/mnt/MORODATA/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 251.712620 secs (426574490 bytes/sec) - 406.81 MB/sec

Supermicro X8DTH-6F-0 motherboard
(1) Xeon E5606 Westmere-EP 2.13GHz Quad
48Gig DDR3 @ 1066
(16) 2TB Seagate Consumer 7200RPM drives
1 Zpool of 4vdevs of 4 disks each in Z2

Stephen
 

RichR

Explorer
Joined
Oct 20, 2011
Messages
77
[pacs@pod2 /]$ dd if=/dev/zero of=/mnt/Store4/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 1224.831318 secs (87664465 bytes/sec)

[pacs@pod2 /]$ dd if=/mnt/Store4/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 631.636189 secs (169993715 bytes/sec)

This does not seem to be very good....

Supermicro X8SIL-F
Intel i3 540 @ 3.07GHz
16GB DDR3-1333
(For this test) 7x3TB Hitachi 5400 Deskstar
Overall 45 3TB Hitachi 5400 Deskstar, 3 Volumes of 7 drives + 3 Volumes of 8 drives

Rich
 

trnagle

Cadet
Joined
Jan 3, 2012
Messages
3
[root@freenas] /mnt/LuckyPool# dd if=/dev/zero of=/mnt/LuckyPool/test.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 120.196348 secs (893323167 bytes/sec)
[root@freenas] /mnt/LuckyPool# dd if=/mnt/LuckyPool/test.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 196.105541 secs (547532629 bytes/sec)

----------

SuperMicro X8DTH-6F-O motherboard
(2) Intel Westmere 6C E5645 2.4G
96GB DDR3 1333
(36) 2TB Toshiba Enterprise 7200 RPM SATAII
1 big raidz2 pool - plan to separate into multiple vdevs

Performance not too bad but I get tons of SCSI status errors on the read, I assume that is why it is so much slower. Any thoughts on tweaks? Any suggestions are welcome! Thanks
 

Ultfris101

Cadet
Joined
Feb 20, 2012
Messages
7
Figured I'd add some numbers. One note, i noticed today that while all of the SATA II drives are correctly identified as such FreeBSD seems to be identifying two SATA III capable drives as SATA I devices at boot time. Have to see if that can be fixed in BIOS or something and if it changes these numbers at all (which it looks like it should):

Specs:

Asus M4A88T-M/USB3 (6 x SATA II)
G.SKILL Value Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1333 (PC3 10600)
AMD Athlon II X3 460
6 x SATA disks, mix of drives ( 3 x 500gb, 1 x 750gb, 2 x 1tb )
ZFS raidz2 volume
FreeNAS-8.0.4-RELEASE-x64 (10351)

dd results:

[root@freenas] /mnt/vol01/music01# dd if=/dev/zero of=tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 672.342579 secs (159701595 bytes/sec)

[root@freenas] /mnt/vol01/music01# dd if=tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 517.364271 secs (207540776 bytes/sec)
 

Stephen J

Dabbler
Joined
Feb 3, 2012
Messages
49
Okay, I made some modifications and re-ran the tests in two configurations. For both, I removed all but 8Gig of RAM from the machine to minimize any caching that might have been going on.

Setup 1: 4 vdevs of 4 drives in RaidZ2

dd if=/dev/zero of=/mnt/MORODATA/tmp.dd bs=2048k count=50k
107374182400 bytes transferred in 758.309176 secs (141596839 bytes/sec) 135MB/sec

dd if=/mnt/MORODATA/tmp.dd of=/dev/null bs=2048k count=50k
107374182400 bytes transferred in 246.927663 secs (434840637 bytes/sec) 414MB/sec

Setup 2: 5 vdevs of 3 drives in 3-way mirror configuration

dd if=/dev/zero of=/mnt/MORODATA/tmp.dd bs=2048k count=50k
107374182400 bytes transferred in 231.364835 secs (464090329 bytes/sec) 442MB/s

dd if=/mnt/MORODATA/tmp.dd of=/dev/null bs=2048k count=50k
107374182400 bytes transferred in 97.432558 secs (1102035960 bytes/sec) 1.02GB/s

Capacity of the pool drops from 14.2TB to 8.9TB but there is a huge speed gain. Definitely enough there to saturate 4 GigE links I would think.

Stephen
 

Hexland

Contributor
Joined
Jan 17, 2012
Messages
110
Just got a new HP Microserver N40L set up... I've been playing with various installs, formatting, etc, and gotten the following results

HP Microserver N40L
8Gb RAM
1 x 2Tb WD EARS
2 x 2Tb Hitachi 7200
1 x 2Tb Seagate Green 5400
Flashed Russian BIOS to enable AHCI on ODD/eSATA ports
1 x Intel PCIe Gigabit Desktop low profile NIC
1 x Koutech IO-PESA111 SATA/eSATA controller

Code:
OpenIndiana 151a
-----------------
  5 x 2Tb Drives (ZFS-RaidZ2), formatted 512K sectors (v.151a has no support for 4k sectors that I can find)
* -----------------------------------------------------------

  dd if=/dev/zero of=/tank/test.dat bs=2048k count=50k
  64.622 seconds (140427796 bytes/sec) -- 133.922Mb/s

  dd if=/tank/test.dat of=/dev/null bs=2048k count=50k
  360.901043 secs (297516963 bytes/sec) -- 283Mb/s



Code:
FreeBSD 9.0
-----------------
  5 x 2TB Drives  (ZFS-RaidZ2), formatted 4096K (manually using gnop create -S 4096)
  -----------------------------------------------------------

  dd if=/dev/zero of=/tank/test.dat bs=2048k count=50k
  555.544 secs (193277531 bytes/sec) -- 184.323Mb/s

  dd if=/tank/test.dat of=/dev/null bs=2048k count=50k
  373.903001 secs (287171224 bytes/sec) -- 273.867Mb/s



Code:
FreeNAS 8.2.0-BETA-2
----------------------
  5 x 2Tb Drives (ZFS-RaidZ2), formatted 512K sectors
  -----------------------------------------------------------
  
  dd if=/dev/zero of=ZFSTank01/test.dat bs=2048k count=50k
  1354.490325 secs (79272757 bytes/sec) -- 75.600Mb/s

  dd if=/tank/test.dat of=/dev/null bs=2048k count=50k
  360.901043 secs (297516963 bytes/sec) -- 283Mb/s



  5 x 2TB Drives  (ZFS-RaidZ2), formatted 4096K (manually using gnop create -S 4096)
  -----------------------------------------------------------

  dd if=/dev/zero of=/tank/test.dat bs=2048k count=50k
  587.706765 secs (182700266 bytes/sec) -- 174.23655Mb/s

  dd if=/tank/test.dat of=/dev/null bs=2048k count=50k
  352.861012 seconds (304295966 bytes/sec) -- 290.199Mb/s


  5 x 2TB Drives  (ZFS Stripe), formatted 4096K (manually using gnop create -S 4096)
  -----------------------------------------------------------

  dd if=/dev/zero of=/mnt/tank/test.dat bs=2048k count=50k  
  353.574893 secs (303681581 bytes/sec) -- 289.61Mb/s



Adding a 60Gb SSD as a cache drive didn't really make that much of a difference in these tests


All I can surmise is that
1. DON'T format 4k sector drives as 512Kb
2. ZFS-RaidZ2 really hits the write performance (I understand it's doing a lot during the write phase, but 100Mb/s performance degradation!?)
3. FreeNAS performs just as well as FreeBSD (as expected), and better than OpenSolaris/OpenIndiana (if OS/Oi get their asses in gear and provide a patched zpool command for 4k formatting, this might change)
 

wussy

Explorer
Joined
Jan 11, 2012
Messages
50
Hi,
Here my performance datas:

Write:
dd if=/dev/zero of=/mnt/store/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 164.120493 secs (654239946 bytes/sec)

Write 2:
dd if=/dev/random of=/mnt/store/tmp.dat bs=512k count=20480 conv=sync
20480+0 records in
20480+0 records out
10737418240 bytes transferred in 170.497767 secs (62976885 bytes/sec)

Read:
dd if=/mnt/store/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 161.410506 secs (665224248 bytes/sec)

Read 2:
dd if=/mnt/store/tmp.dat of=/dev/zero bs=512k count=20480 conv=sync
20480+0 records in
20480+0 records out
10737418240 bytes transferred in 3.290633 secs (3263025187 bytes/sec)


Config:
FreeNAS Version FreeNAS-8.0.4-RELEASE-x64 (10351)
Plattform 2x Intel(R) Xeon(R) CPU E5607 @ 2.27GHz
Speicher 65461MB
Adaptec Raid 52445

RaidZ2 24x 3TB (one spare) with 24 3TB Barracuda 7200 RPM 64MB 6GB/s (ST300DM001)
 

Hexland

Contributor
Joined
Jan 17, 2012
Messages
110
FreeNAS Version FreeNAS-8.0.4-RELEASE-x64 (10351)
Plattform 2x Intel(R) Xeon(R) CPU E5607 @ 2.27GHz
Speicher 65461MB
Adaptec Raid 52445

RaidZ2 24x 3TB (one spare) with 24 3TB Barracuda 7200 RPM 64MB 6GB/s (ST300DM001)

Holy crap dude! I hate you ;)
 

paleoN

Wizard
Joined
Apr 22, 2012
Messages
1,403
FreeNAS-8.2.0-BETA3-x64
SilverStone Kublai SST-KL04B
ASUS F1A75-V Pro
AMD A6-3500 Llano CPU
8GB DDR3 RAM
Mirror 4k forced, 2 x Seagate ST2000DM001 2TB

Write:
Code:
# dd if=/dev/zero of=/mnt/tank/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 611.257671 secs (175661080 bytes/sec)


Read:
Code:
# dd if=/mnt/tank/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 410.381584 secs (261644739 bytes/sec)

The mirror has overhead on writing but faster on reading as advertised.


For fun-

512 byte block size:
Code:
# dd if=/dev/zero of=/mnt/tank/tmp.dat bs=512 count=204800k
209715200+0 records in
209715200+0 records out
107374182400 bytes transferred in 2262.562246 secs (47456897 bytes/sec)

# dd if=/mnt/tank/tmp.dat of=/dev/null bs=512 count=204800k
209715200+0 records in
209715200+0 records out
107374182400 bytes transferred in 889.157625 secs (120759446 bytes/sec)

Good example of what the OP was talking about.


4K block size:
Code:
# dd if=/dev/zero of=/mnt/tank/tmp.dat bs=4096 count=25600k conv=sync
26214400+0 records in
26214400+0 records out
107374182400 bytes transferred in 758.922298 secs (141482445 bytes/sec)

# dd if=/mnt/tank/tmp.dat of=/dev/null bs=4096 count=25600k conv=sync
26214400+0 records in
26214400+0 records out
107374182400 bytes transferred in 416.958528 secs (257517655 bytes/sec)

Still significantly slower on writes.


Raw before I created the mirror:
This will do terrible things to your array. Do not try it on a disk in any type of array.
Code:
# dd if=/dev/zero of=/dev/ada0 bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 543.106874 secs (197703597 bytes/sec)

# dd if=/dev/ada0 of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 537.182353 secs (199884046 bytes/sec)
 

NetCaptive

Cadet
Joined
Apr 30, 2012
Messages
3
Here's my setup:
HP Microserver N40L
8Gb RAM
5 x 3Tb WD 30EZRX
Flashed BIOS with bios-mods BIOS (see http://www.bios-mods.com/forum/Thread-HP-Proliant-Microserver-AMI-BIOS-MOD)
1 x LSI Logic SAS9220-8i SAS Controller Flashed to SAS9211-IT mode (see http://www.servethehome.com/ibm-serveraid-m1015-part-4)
0 x Onboard ports used (all drives are on the above controller).
1 x ZFS-RAIDZ Pool

Freenas 8.20BETA2:
Write:
Code:
[root@freenas] /mnt/pool_0# dd if=/dev/zero of=/mnt/pool_0/tmp.dat bs=2048k count=50k 
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 409.302018 secs (262334847 bytes/sec)

Read:
Code:
[root@freenas] /mnt/pool_0# dd if=/mnt/pool_0/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 332.720384 secs (322715973 bytes/sec)


Freenas 8.20BETA3:
Write:
Code:
[root@freenas] /mnt/pool_0# dd if=/dev/zero of=/mnt/pool_0/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 569.011866 secs (188702888 bytes/sec)

Read:
Code:
[root@freenas] /mnt/pool_0# dd if=/mnt/pool_0/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 361.055395 secs (297389774 bytes/sec)


Not sure why there's such a big discrepancy between versions. It's an unused, freshly built system. I did upgrade from beta2 to beta3 and rebooted, that's all.
 

geister

Cadet
Joined
May 15, 2012
Messages
3
Hello everyone.
These are the results I am getting:

[root@freenas] ~# dd if=/dev/zero of=/mnt/RaidZ/tmp.dat bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 66.124683 secs (1623813948 bytes/sec)

[root@freenas] ~# dd if=/mnt/RaidZ/tmp.dat of=/dev/null bs=2048k count=50k
51200+0 records in
51200+0 records out
107374182400 bytes transferred in 46.225716 secs (2322823559 bytes/sec)

It works out to some crazy number:
Write: 1548.58 MB/s
Read: 2215.21 MB/s


System
CPU: AMD Phenom(tm) II X2 550 Processor Black Edition(3120.89-MHz K8-class CPU)
Memory: 16GB Kingston DDR3 1600mhz
Boot drive: Corsair Voyager 32 GB
Storage: 5 x 1TB Samsung Spinpoint F4
Controller Standard Onboard SATA (GIGABYTE GA-970A-UD3)
File System: ZFS RaidZ
Network: Intel Pro 1000PT Dual Port
 
Top