Linux Client Performance - CIFS and NFS

A CentOS 6.2 virtual machine was used to evaluate NFS and CIFS performance of the NAS when accessed from a Linux client. We chose IOZone as the benchmark for this case. In order to standardize the testing across multiple NAS units, we mount the CIFS and NFS shares during startup with the following /etc/fstab entries.

//<NAS_IP>/PATH_TO_SMB_SHARE /PATH_TO_LOCAL_MOUNT_FOLDER cifs rw,username=guest,password= 0 0

<NAS_IP>:/PATH_TO_NFS_SHARE /PATH_TO_LOCAL_MOUNT_FOLDER nfs rw,relatime,vers=3,rsize=32768,wsize=32768,namlen=255,hard,proto=tcp,timeo=600,retrans=2, sec=sys,mountaddr <NAS_IP>,mountvers=3,mountproto=udp,local_lock=none,addr=<NAS_IP> 0 0

The following IOZone command was used to benchmark the CIFS share:


IOZone provides benchmark numbers for a multitude of access scenarios with varying file sizes and record lengths. Some of these are very susceptible to caching effects on the client side. This is evident in some of the graphs in the gallery below.

Readers interested in the hard numbers can refer to the CSV program output here.

The NFS share was also benchmarked in a similar manner with the following command:

IOZone -aczR -g 2097152 -U /nfs_test_mount/ -f /nfs_test_mount/testfile -b <NAS_NAME>_NFS_EXCEL_BIN.xls > <NAS_NAME>_NFS_CSV.csv

The IOZone CSV output can be found here for those interested in the exact numbers.

A summary of the bandwidth numbers for various tests averaged across all file and record sizes is provided in the table below. As noted previously, some of these numbers are skewed by caching effects. A reference to the actual CSV outputs linked above make the entries affected by this effect obvious.

QNAP TS-451 - Linux Client Performance (MBps)
Init Write 67 69
Re-Write 69 74
Read 34 125
Re-Read 34 125
Random Read 21 62
Random Write 62 71
Backward Read 21 49
Record Re-Write 788* 1317*
Stride Read 32 106
File Write 68 79
File Re-Write 69 80
File Read 24 89
File Re-Read 24 92
*: Number skewed by caching effect


Single Client Performance - iSCSI on Windows Encryption Support Evaluation
Comments Locked


View All Comments

  • BMNify - Monday, July 28, 2014 - link

    just to add "the hardware you're suggesting looks like DIY enterprise architecture" its also possible to actually buy NICE looking and cheap Free Standing Rack Cabinets for the home too now such as the Orion Free Standing Rack Cabinets with glass front for one

    a 9U glass fronted cabinet for £240.00 (inc VAT) to sit next to your desk in the SOHO room or a cupboard if your not into showing off your home made rack... :)
  • Trickie - Tuesday, July 29, 2014 - link

    Came across this when researching the x51 series last week. A x53 model is in the pipeline with a j1900 chip. Guessing they will be targeting small business with this model and be charging an even bigger premium as a result even though the chip cost $10 more.
    The extra grunt will suit my use case much better (vm's). What I need to make the jump to a NAS from a desktop is hardware transcoding support within plex. See
  • azazel1024 - Monday, July 28, 2014 - link

    That is interesting that the J1800 doesn't support AES-NI instructions. I was going to post a self righteous comment about "well of course it does!", and then I checked Intel ARK and saw they list it as not supporting it. My humble little z3740 in my tablet DOES support AES-NI though.

    I always find it odd what Intel choose to enable and disable on their various SKUs *shakes head slowly*
  • takeshi7 - Monday, July 28, 2014 - link

    You should review the Seagate NAS Pro 4-bay next. I'm curious how the Intel based Seagate compares to this QNAP.
  • halfflat - Monday, July 28, 2014 - link

    Still no ECC RAM? Can't really take it seriously.
  • ganeshts - Monday, July 28, 2014 - link

    Look at the target market : Home media enthusiasts / power users - who want to stream their huge media collections / backup their smartphone photos / need a backup for small amounts of data that they generate on their laptops (say, tax returns or documents or similar things). Why go in for ECC RAM overkill (and associated increased platform cost?)

    ECC RAM is necessary only for mission-critical applications. If you feel ECC RAM is necessary for a non-ZFS mdadm-based software RAID system like the TS-451, I would love to hear the arguments in its favor.
  • darkfalz - Monday, July 28, 2014 - link

    Would prefer a 6 bay. At 25% space penalty, RAID-5 with 4 drives is a bit painful. With 6 bays you're down to at more acceptable 17% space penalty. Give me 6x6TB in RAID-5 NAS and I'll be happy (currenting running 5x4TB in a HTPC but feeling cramped already!)
  • wintermute000 - Tuesday, July 29, 2014 - link

    there is a 6 bay model, its just not the one reviewed lol
  • basroil - Monday, July 28, 2014 - link

    Those are some really nice numbers for iSCSI (which you need for Lightroom to work), far better than other qnap devices...

    Really got to say Anandtech rocks, the really do listen to feedback and test for cases that readers are interested in! (hell, I like the service so much I disabled all adblock like scripts for the site, something I never do)
  • - Tuesday, July 29, 2014 - link

    759$ - overpriced

Log in

Don't have an account? Sign up now