Single Client Performance - CIFS & iSCSI on Windows

The single client CIFS and iSCSI performance of the Asustor AS6204T was evaluated on the Windows platforms using Intel NASPT and our standard robocopy benchmark. This was run from one of the virtual machines in our NAS testbed. All data for the robocopy benchmark on the client side was put in a RAM disk (created using OSFMount) to ensure that the client's storage system shortcomings wouldn't affect the benchmark results. It must be noted that all the shares / iSCSI LUNs are created in a RAID-5 volume.

The Asustor AS6204T provides acceptable performance in all workloads without particularly standing out in any particular one. That said, it manages to be in the top half of the performance pack in almost all of them.

HD Video Playback - CIFS

2x HD Playback - CIFS

4x HD Playback - CIFS

HD Video Record - CIFS

HD Playback and Record - CIFS

Content Creation - CIFS

Office Productivity - CIFS

File Copy to NAS - CIFS

File Copy from NAS - CIFS

Dir Copy to NAS - CIFS

Dir Copy from NAS - CIFS

Photo Album - CIFS

robocopy (Write to NAS) - CIFS

robocopy (Read from NAS) - CIFS

We created a 250 GB iSCSI LUN / target and mapped it on to a Windows VM in our testbed. The same NASPT benchmarks were run and the results are presented below. The iSCSI performance is slightly better than CIFS performance, but, again, nothing to make the unit stand out in the crowd.

HD Video Playback - iSCSI

2x HD Playback - iSCSI

4x HD Playback - iSCSI

HD Video Record - iSCSI

HD Playback and Record - iSCSI

Content Creation - iSCSI

Office Productivity - iSCSI

File Copy to NAS - iSCSI

File Copy from NAS - iSCSI

Dir Copy to NAS - iSCSI

Dir Copy from NAS - iSCSI

Photo Album - iSCSI

robocopy (Write to NAS) - iSCSI

robocopy (Read from NAS) - iSCSI

Both CIFS and iSCSI have scope for performance improvements. It has to be kept in mind that ADM (the Asustor OS) is still a fledgling compared to other mature operating systems such as QNAP's QTS and Synology's DSM. The important takeaway here is that the system provides acceptable performance for all the workloads.

Introduction and Testbed Setup Encryption Support Evaluation


View All Comments

  • extide - Thursday, November 5, 2015 - link

    I think Purch could handle that expense. Anandtech isn't a little teensy website anymore, remember. Reply
  • DanNeely - Thursday, November 5, 2015 - link

    Unless your preferred solution is either the Toms Hardware (picked on only because I'm not sure what else Purch already owns) comment system reskinned to Anandtech colors or something completely off the shelf like Discuss; that might not be the best idea. I'm very leery of handing the job off to a corporate overlord who bought the site on the assumption of being able to pull extra money out of it and which has already triggered multiple complaints from readers about overly intrusive advertising. Reply
  • milkod2001 - Thursday, November 5, 2015 - link

    You could possibly contact Purch:
    (It own this website) and request it there. For such reputable company it is shame there is not edit option. It should not take more then 1 day to implemented it or just install DISQUS
  • Namisecond - Monday, December 7, 2015 - link

    Don't underestimate the time it may take to transfer all the user accounts over. If the tools work fine and everything goes well in a single pass, it could take a day. If there are serious problems, might take several days to over a week. Reply
  • extide - Thursday, November 5, 2015 - link

    Perhaps it refers to the number of dual-core modules -- as that is how Intel builds those chips (in dual-core pairs) Reply
  • dtgoodwin - Thursday, November 5, 2015 - link

    Please, please, could you review a moderately configured FreeNAS setup testing it the same way you do these? I know there are a tremendous amount of variables, and tuning available, but if you used the standard configuration of software, and chose a medium performance platform such as an i3, or core series Pentium or Celeron and tested it with 4 drives, it could be comparable.

    I run a 7 drive RAIDZ2 with 3TB WD REDs. I am almost always able to saturate a single GB link using CIFS and robocopy or Windows file copy whether reading or writing. I'm running a less than optimal setup as I have 15 TB of usable space, but only 8GB of RAM. My system with a SuperMicro board with ICMP, 2 GB NICs, a Pentium G2030, and a used SuperMicro case cost me about $650 not including the drives.
  • Black Obsidian - Thursday, November 5, 2015 - link

    I'm sure they could write such a feature, but the first 17 pages would be component selection, installation, lexicon (vdev, pool, stripe size, etc.), Shell Commands 101, and so forth.

    And none of it would say anything new. Anyone running--or considering running--FreeNAS knows its performance capabilities, and anybody seriously considering a Synology/QNAP/Asustor product is going to be in a coma by the third page of such a feature.
  • DanNeely - Thursday, November 5, 2015 - link

    They wouldn't have to write those pages if they didn't want to. Just set general requirements and ask the hardware partners to donate a build as usual; or punt farther and get an entry level prebuilt freenas box from that company that's been selling them for years (albeit at a price that makes Synology/QNAP boxes look cheap). Default to RAIDZ5 to maintain parity with the rest of their 4 bay NAS reviews. Stick to setting it up via the installer/web config panel. The whole point of a product like freenas vs roll your own is that for typical nas use cases the user should never need to drop down to the shell or need to understand all of the magic behind the scenes. A single page on what ZFS does better than ETX4 on most linux based nases is really all that's needed. Reply
  • Navvie - Friday, November 6, 2015 - link

    HP Microserver G1620T. Upgrade RAM to 8GB. Hardware selection done.
    FreeNAS/nas4free on USB stick. Default ZFS options that the web interface presents.

    AnandTech seems to be making a big thing of NAS devices these days, ignoring the build your own device option is silly. Even if somebody is set on buying an off the shelf unit, it would be nice to see how much better it performs (or not) compared to a BYOD.
  • DCide - Thursday, November 5, 2015 - link

    For some reason Asustor always reminds me of the words stupid or stupor. Probably not ASUS' finest branding moment. Reply

Log in

Don't have an account? Sign up now