Supermicro X7SPA L/H/HF ATOM serverboards (Level 1 Tested)


Recommended Posts

  • Replies 397
  • Created
  • Last Reply

Top Posters In This Topic

I've read the atom is as fast as a PIII 900. The numbers posted do not jive with that value. Which is good news!

 

From what I have read about the Atom processor they had always compared it to and said it was at about the level of a Pentium M

 

Which Pentium M?

I would surmise a 1.7 or so, but just curious. It was the PIII article which sort of scared me away from the Atom, but in relation to a Pentium M and now with a decent set of benchmarks I have renewed faith in this processor line.

Link to comment

Which Pentium M?

I would surmise a 1.7 or so, but just curious. It was the PIII article which sort of scared me away from the Atom, but in relation to a Pentium M and now with a decent set of benchmarks I have renewed faith in this processor line.

 

A quick Google Search lead me to this article which says it is near a Pentium M Dothan 90nm processor.  I think I read the same thing in another article but can't find it right now.

Link to comment

The Bogomips was just for my own education where the processor will be if I replace one of my machines with this board. The current machine is a dual xeon lv 2.4 rig that is a firewall, proxy, IRC bot ana vmware server for ripping/transcoding. I'm just deciding if I want to give up some processing power or not. So the number is helpful in revealing the cpu is faster then an article I read which was incorrect.

 

I found the following links helpful too.

http://www.cpubenchmark.net/cpu_list.php

http://www.cpubenchmark.net/cpu_lookup.php?cpu=Intel+Atom+330+%40+1.60GHz

 

Link to comment

Does anyone with this board have a Kill-a-watt load meter they can plug their server into to get idle power measurements?

 

I am considering buying one but cannot find power draw data for this board anywhere.  I saw the measurements posted earlier in this thread and it sounds like they are suspect.  

 

If you can do this, please measure with all hard drives and peripherals disconnected from the power supply and let us know what power supply you are using since power supply efficiency will impact the result.

 

Thanks in advance!

 

Just for reference, I have an inexpensive ECS 945GCD-M Dual Core Atom 330 board here and it measures < 38W at idle.  That's with (2) hard drives plugged in (but spun down).  Taking power supply efficiency into account and the standby wattage of the drives, the board was only consuming (38*.75)-1 =  28W.  

 

I would expect much better from this pineview platform but want to understand how much the extra GPU is hurting this system (it also has an on-package GPU which is not being used, but may still be powered).

Link to comment

And for reference, some of us have even more powerful low-power Intel Dual Core CPU systems with 4 drives which idle less than this board alone does.

Not to derail this thread, but can you comment on this further, or point to/start a thread were you may have already elaborated on this? You've piqued my interest.
Link to comment

I have C2D E7200 now. Consumption is bellow line of sign - 45W in IDLE and with 6xSATA HDD spun down.

 

Write throughput to UnRaid from Win XP is 40MB/s.

Parity check is avarage 65-75MB/s. Disk capacity is 95% full. Time of parity check is 4 hours now.

 

I am earlier should have mobo with Pentium M 760. Consumption was about 43W with 6xSATA HDD spun down.

Link to comment

 

Write throughput to UnRaid from Win XP is 40MB/s.

Parity check is avarage 65-75MB/s. Disk capacity is 95% full. Time of parity check is 4 hours now.

 

 

How are you measuring your write throughput?  Is this with a cache drive?

 

No. I do not use cache drive now. I am copy through SAMBA from WinXP through integrated addon TeraCopy into Explorer. TeraCopy show write and if you check Test then show read speed.

Link to comment

No. I do not use cache drive now. I am copy through SAMBA from WinXP through integrated addon TeraCopy into Explorer. TeraCopy show write and if you check Test then show read speed.

 

I'm not sure what that software is doing, but it claims to allocate buffers and optimize transfer speeds through some software tweaks.  

 

It would interesting for you to run this test from a command prompt and see what transfer speeds you get.  The below program IOZone attempts to get at raw transfer performance without operating system and other software optimizations getting in the way:

 

Free download here:  http://www.iozone.org/

 

Browse to your installation location on the client machine from a command prompt and issue the following command:

 

iozone -Rab RESULTS.xls -i 0 -i 1 -+u -f \\[iNSERT SERVER NAME]\disk1\filetest -y 64k -q 64k -n 64k -g 4G -z

 

This will automatically create an excel spreadsheet called RESULTS.xls in the local directory where you ran the benchmark where it will capture all of the read and write throughput numbers.  I have been quoting only the 4GB numbers for read and write (not re-read or re-write).

 

There is a large database of what other commercial NAS systems achieve on this website for comparison using the same arguments as above:

 

http://www.smallnetbuilder.com/component/option,com_nas/Itemid,190

Link to comment

If I am write to Unraid from Windows 7 ultimate from Explorer then show 32-37MB/s

 

I can do test again.

 

Domain www.smallnetbuilder.com I know.

 

I am attach another process of copy from WindXP to Unraid. Green bars is write to server, red bars is read from server. Peek is 52 MB/s.

copy-to-ICH10-E7200.JPG.3c010e7941c8d58fa323e009d9c70403.JPG

Link to comment

If I am write to Unraid from Windows 7 ultimate from Explorer then show 32-37MB/s

 

I can do test again.

 

Domain www.smallnetbuilder.com I know.

 

I am attach another process of copy from WindXP to Unraid. Green bars is write to server, red bars is read from server. Peek is 52 MB/s.

Very nice "write" performance.  Can't wait until the unRAID kernel is upgraded to 2.6.32.7 to see how your server performs.  It just might pick up another 20% in write speed as mentioned here: http://lime-technology.com/forum/index.php?topic=5146.msg48413#msg48413

 

Joe L.

Link to comment

Very nice "write" performance.  Can't wait until the unRAID kernel is upgraded to 2.6.32.7 to see how your server performs.  It just might pick up another 20% in write speed as mentioned here

 

Joe L., so that this performance is because of new kernel version in UnRaid 4.5.1 ?

 

unRAID 4.5.1 still uses the 2.6.31 series of Linux kernels. The 2.6.32 series switched the way the system handles flushing of data for the better. From the initial tests performed, it looks to be around 20% improvement for network writes. Future unRAID versions should make use of the 2.6.32 series, it's a matter of when.

 

The Linux kernel developers indicated that 2.6.32 will be the "long maintained" versions of the kernel. What this means is it will be maintained for a 2 - 3 year duration. [ http://thread.gmane.org/gmane.linux.kernel/939800 ]

 

2.6.31-stable

 

Today the last 2.6.31-stable kernel was released, all users of this

kernel series are strongly encouraged to switch to the 2.6.32 kernel

series, as there will not be any more updates for this branch in the

future.

 

2.6.32-stable

 

I'd like to announce that the 2.6.32-stable tree is also going to be

maintained as a "long-term" stable release, living for 2-3 years, like

the 2.6.27 kernel is.  This is because a number (i.e. more than 2) Linux

distributions are basing their "enterprise" releases on this kernel

version, and it will make their lives easier if I keep it alive.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.