Like every other website on the planet, SmallNetBuilder uses cookies. Our cookies track login status, but we only allow admins to log in anyway, so those don't apply to you. Any other cookies you pick up during your visit come from advertisers, which we don't control.
If you continue to use the site, you agree to tolerate our use of cookies. Thank you!

Wi-Fi Router Charts

Click for Wi-Fi Router Charts

Mesh System Charts

Click for Wi-Fi Mesh System Charts


The N6850 was tested with 2.02.04 firmware using our NAS test process.

The Benchmark summary below shows pretty consistent Windows File Copy performance for RAID 0, 5 and 10 modes at 107 MB/s, with 104 and 103 MB/s, respectively. File Copy read is also consistent at 106, 105 and 100 MB/s for RAID 0, 5 and 10, respectively.

NASPT File copy results are higher than their Windows File Copy counterparts for write (126, 126, 129 MB/s) , but lower for read (96, 8987, 97 MB/s) for RAID 0, 5 and 10. Note that the writes are higher than 125 MB/s, which is the limit for a Gigabit Ethernet connection. So, clearly, the N6850 is being held back by its Gigabit Ethernet connection.

Thecus N6850 NAS Benchmark summary
Thecus N6850 NAS Benchmark summary

iSCSI target write performance to target created on a RAID 5 volume came in at 95 MB/s, with read a tad lower at 93 MB/s.

I downloaded and installed Thecus' USB / eSATA Schedule Backup module to run backup tests with our standard Iomega UltraMax Pro Desktop Hard Drive for USB 2.0 and eSATA connections. I also connected a WD MyBook 3.0 to run USB 3.0 backup tests.

Since Thecus doesn't provide a way to format external drives, I was able to test only FAT32 and NTFS backup formats. Best backup throughput of 84 MB/s was obtained with the WD USB 3.0 drive NTFS formatted, while the slowest 28 MB/s speed was to the Iomega drive connected via USB 2.0 with FAT32 format.

Rsync backup to the NAS Testbed running DeltaCopy acting as an rsync target came in at a comparatively poky 39 MB/s. Note that, as is our standard practice, this is without compression or encryption options enabled.

For a competitive look, I decided to use the NASPT RAID 10 File copy benchmark. The chart below shows all products in the NAS chart database, regardless of number of bays. This is a fair comparison because our standard practice is to use only four-drive volumes, even on NASes that have more than four bays.

The only non-Atom Intel-powered NASes in the chart below is the QNAP TS-1079 Pro (i3-2120). The QNAP TS-559 Pro II and Synology DS1512+ achieve their high performance with dual-core D2700 and D525 Atoms instead.

Intel NASPT RAID10 File Copy To NAS
Intel NASPT RAID10 File Copy To NAS

Rankings shift around a bit for NASPT RAID10 read, with the N6850 moving to the top spot.

Intel NASPT RAID10 File Copy From NAS
Intel NASPT RAID10 File Copy From NAS

Use the NAS Charts to further explore performance.

10 GbE Test

Updated 7/27/2012

I haven't done much testing using a 10 GbE interface because most NASes haven't been powerful enough to take advantage of it. But it looks like I may need to start testing 10 GbE performance on a more regular basis, or at least with two aggregated Gigabit ports.

Thecus was kind enough to send two Intel X520-DA2 10 GbE boards with the N6850 and a connecting cable so that I could do just that. One board went into the PCIe x8 slot in the NAS, the other into my NAS testbed in the ASUS P7H57D-V EVO motherboard's PCIe 2.0 x 16 5GT/s slot. I downloaded the latest driver ( Win 7 64 bit), set static IP addresses on the testbed and NAS 10 GbE ports, connected the cable and disconnected the Gigabit Ethernet cable to ensure that the 10 GbE interface was used. I then ran our standard benchmark suite, which runs the Windows file copy and Intel NASPT tests three times. The NAS was configured in RAID 5 for the test.

The results are summarized in the chart below and compared to the same tests run on a QNAP TS-1079 Pro. The QNAP results differ a bit from those published in its review because QNAP recently sent the NAS back in for a 10 GbE retest with an Emulex adapter. Unfortunately, the Emulex adapter sent required Win 7 Pro and the NAS testbed has only Home Premium installed. So I used the Intel X520 to run another test instead, the results of which are shown below.

Thecus N6850 vs. QNAP TS-1079 Pro - 10 GbE connection
Thecus N6850 vs. QNAP TS-1079 Pro - 10 GbE connection

The results show that the N6850 edges out the QNAP in every test except the simultaneous playback of four HD videos (HDVideo_4Play). I have no idea why this would be, particularly because the single HD video playback test right above it favors the Thecus by a wide margin.

I'm including the Windows filecopy and iSCSI tests, but don't put too much faith in them. These tests all write and read large files from the RAID 0 array on the testbed, which is capable of only 200 MB/s writes and reads. The Intel NASPT tests, on the other hand, run directly from memory, so are not disk I/O bound.

It's also possible that the i3-540 based testbed could also be limiting performance. But at least it is applying the same limit to both products. Looks like it may soon be time to upgrade the NAS testbed again!

Closing Thoughts

The N6850 sticks closely to Thecus' playbook. It's a bit rough around the edges software-wise, but doesn't hold back when it comes to performance. Also standard, and perhaps to make prospective high-performance NAS buyers take a serious look, is the N6850's very aggressive price. While the cheapest you can get a TS-1079 Pro is around $2400, the N6850 will cost you half of that. So with the money you save on the N6850, you can pick up two 10 GbE cards and still come out ahead.

Support Us!

If you like what we do and want to thank us, just buy something on Amazon. We'll get a small commission on anything you buy. Thanks!

Don't Miss These

  • 1
  • 2