hello
i want to add the other benchmark results with same config and sas drives:

mainboard: p5000, 8core xeon, 4 gb ram, 2 x sas lsi-1068 controller
wd drives: 2 tb raid edition
sas drives: 15K/m seagate 3,5". 146 GB

wd 2TB x 7, z3, dedup and compress on, no ssd
daten   12.6T       start   2010.02.17      8G      202 MB/s    83      10 MB/s     4   4.436 MB/s      5   135 MB/s    87      761 MB/s

sas 15k, 146GB x 4, z3+dedup and compress off, no ssd:
z3nocache   544G    start   2010.02.18      8G      71 MB/s     31      84 MB/s     15      47 MB/s     13      87 MB/s     55      113 MB/s    

sas 15k, 146GB x 4, z3+dedup and compress on, no ssd:
z3nocache   544G    start   2010.02.18      8G      218 MB/s    99      410 MB/s    92      171 MB/s    50      148 MB/s    92      578 MB/s 

sas 15k, 146GB x 4, z3+dedup and compress on + ssd read cache:
z3cache     544G    start   2010.02.17      8G      172 MB/s    77      205 MB/s    40      95 MB/s     27      141 MB/s    90      546 MB/s  


##################### result ##################################
all pools are zfs z3
sas are Seagate 15K/m drives, 146 GB


                        seq-write-ch  seq-write-block   rewrite        read-char    read-block

wd 2gb x7                202 MB/s      10 MB/s           4,4 MB/s       135 MB/s     761 MB/s

sas 15k x 4 no dedup     71 MB/s      84 MB/s           47 MB/s         87 MB/s     113 MB/s                           
sas 15k x 4 +dedup+comp 218 MB/s     410 MB/s          171 MB/s        148 MB/s     578 MB/s   
sas 15k x 4 +ded+ssd    172 MB/s     205 MB/s           95 MB/s        141 MB/s     546 MB/s


conclusion:
if you need performance: 
use fast sas drives
activate dedup and compress (if you have enough cpu power)
ssd read cache is not important in bonnie test

high capacity drives are very well in reading and seq. writing

gea

Hi Guenther.

I am just learning how ZFS works.

Thanks for putting up some benchmarks. I am trying to figure out how the different disk combinations perform.

With dedup and compression enabled, what is the CPU load like during the test?

I wonder what the write performance of the 7200 RPM drives would be if SSD was used as a ZIL. Just thinking if you were in a write intensive environment, how much would having a SSD ZIL improve performance.

Have a great day!

Geoff

hello geoff

cpu load is the number behind transfer rate see http://www.napp-it.org/bench.png

but it*s obvious, dedup and compress will improve performance a lot - if you have enough cpu power and ram.

i will add a bonnie benchmark with 7200k drives with/without ssd read cache. i have just no fast ssd*s to test write cache. i will add these tests when i get new ssds based on sandforce controller in march.

gea

hello
i have made some benchmarks with my napp-it zfs-server

all Benchmark results from bonnie++ 1.03:
www.napp-it.org/bench.pdf

-> 2gb vs 4 gb vs 8 gb ram
-> mirror vs raidz vs raidz2 vs raidz3
-> dedup and compress enabled vs disabled

result in short:
8gb ram vs 2 Gb: + 10% .. +500% more power (green drives)
compress and dedup enabled: + 50% .. +300%
mirror vs Raidz: fastest is raidz, slowest mirror, raidz level +/-20%


main-results:

high speed drives Seagate ST3146855SS
-----------------------------------------------
NAME SIZE Bonnie Date(y.m.d) File Seq-Out(Char) %CPU Seq-Out(Block) %CPU Seq-Out(Rewrite) %CPU Seq-In(Char) %CPU Seq-In(Block) %CPU Rnd Seeks %CPU Files Seq-Create Rnd-Create
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
ram 2 vs 4 vs 8 gb
-------------------
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z2, 2 GB RAM, dedup and compress enabled
z2-hd4 544G start 2010.02.19 4G 201 MB/s 98 246 MB/s 47 147 MB/s 48 136 MB/s 98 656 MB/s 95 4181.4/s 22 16 27528/s 19871/s
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z2, 4 GB RAM, dedup and compress enabled
z2-hd4 544G start 2010.02.19 8G 213 MB/s 97 402 MB/s 87 219 MB/s 66 162 MB/s 98 719 MB/s 91 5184.7/s 22 16 22213/s 20805/s
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z2, 8 GB RAM, dedup and compress enabled
z2-hd4 544G start 2010.02.19 16G 224 MB/s 98 285 MB/s 53 262 MB/s 80 167 MB/s 99 813 MB/s 92 4375.9/s 17 16 23405/s 16510/s
dedup and compress enabled vs disabled
--------------------------------------------------------------------------------
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z2, 8 GB RAM, compress and dedup disabled
z2-hd4 544G start 2010.02.19 16G 125 MB/s 65 193 MB/s 57 86 MB/s 26 111 MB/s 71 214 MB/s 29 1269.0/s 23 16 28817/s 19659/s
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z2, 8 GB RAM, dedup and compress enabled
z2-hd4 544G start 2010.02.19 16G 224 MB/s 98 285 MB/s 53 262 MB/s 80 167 MB/s 99 813 MB/s 92 4375.9/s 17 16 23405/s 16510/s
mirror vs raidz vs raidz2 vs raidz3
---------------------------------------------------------------------------------------
2 x SAS Seagate 3,5" 15K/m 146 GB, Mirror, 8 GB RAM, dedup and compress enabled
mi-hd2 136G start 2010.02.19 16G 208 MB/s 98 262 MB/s 49 250 MB/s 73 157 MB/s 93 696 MB/s 72 5654.1/s 39 16 23623/s 17026/s
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z, 8 GB RAM, dedup and compress enabled
z-hd4 544G start 2010.02.19 16G 227 MB/s 98 445 MB/s 90 302 MB/s 82 167 MB/s 99 681 MB/s 75 4924.9/s 28 16 30426/s 18859/s
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z2, 8 GB RAM, dedup and compress enabled
z2-hd4 544G start 2010.02.19 16G 224 MB/s 98 285 MB/s 53 262 MB/s 80 167 MB/s 99 813 MB/s 92 4375.9/s 17 16 23405/s 16510/s
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z3, 8 GB RAM, dedup and compress enabled
z3-hd4 544G start 2010.02.19 16G 222 MB/s 98 418 MB/s 92 217 MB/s 59 151 MB/s 91 624 MB/s 64 3865.9/s 33 16 27983/s 19808/s
with ssd intel g2:
-----------------
4 x SAS Seagate 3,5" 15K/m 146 GB Raid Z2, 8 GB RAM, ssd read cache INTEL SSDSA2M160 160 GB MLC (2nd generation), dedup and compress enabled
z2-hd4 544G start 2010.02.19 16G 218 MB/s 98 404 MB/s 86 271 MB/s 79 163 MB/s 98 649 MB/s 82 3624.9/s 18 16 25852/s 19204/s

gea
www.napp-it.org

hello

napp-it v. 0.222 is out

• bugfix (error add drive to pool)
• detailed hd-info, see screenshot

gea

www.napp-it.org

Hi,

I got a curious problem with set_pw function: Wenn I create a new user or try to change the password of an existing user it seems the password is not really set or changed. There is an entry in the shadow file but I am not able to add such a user to a smb group (SID not found ...). Only wenn I set the password from a commandline with 'passwd ...' the user could be added to a smb group. I big wish: Please could you change all the password inputs to invisible, because they are visible during input and also visible forever in the browser.

Thanks, Ernst

hello ernst

1. are you in workgroup or domain mode?

2. hidden pw on input

you are right.

at the moment all user management in napp-it is only for basic home-use.

i use and suggest ad and do all user management there. so this function was impemented after basic testing. i do not use that regularly.

i will fix these in next beta.

gea

Hi,

It is just a home network with a workgroup.

It is a little bit curious why the chpasswd seems not to work properly.

Ernst