Archive Home arrow Reviews: arrow Network arrow QNAP TS-879U-RP 10GbE NAS Server
QNAP TS-879U-RP 10GbE NAS Server E-mail
Reviews - Featured Reviews: Network
Written by Bruce Normann   
Monday, 19 March 2012
Table of Contents: Page Index
QNAP TS-879U-RP 10GbE NAS Server
Closer Look: 10GbE QNAP TS-879U-RP
QNAP TS-879U-RP NAS Hardware
NAS Testing Methodology
RAID 5 Test Results
Intel NASPT Test Results
ATTO Disk Benchmark Results
NAS System Overhead Measurements
NAS Server Final Thoughts
QNAP TS-879U-RP 10GbE Conclusion

Intel NASPT Test Results

NASPT brings an important perspective to our test protocol, as it is designed to measure the performance of a NAS system, as viewed from the end user's perspective. Benchmarks like ATTO use Direct I/O Access to accurately measure disk performance with minimal influence from the OS and the host platform. This provides important, objective data that can be used to measure raw, physical performance. While it's critical to measure the base performance, it's also important to quantify what you can expect using real-world applications, and that's exactly what NASPT does. One of the disadvantages of NASPT is that it is influenced by the amount of memory installed on the client, and it was designed for systems that had 2-4 GB of RAM. Consequently, two of the tests give unrealistic results, because they are measuring the speed of the buffer on the client, instead of the actual NAS performance. For that reason, we will ignore the results for "HD Video Record" and "File Copy to NAS".

First, let's look at the results with the standard GbE interface, and un-encrypted drives. This is sort of the baseline, as we want to see how much AES-256 encryption reduces the performance, and we want to see how much the enhanced Ethernet connection improves the performance. With the basic GbE interface in place, a familiar pattern is seen. No individual test pushes past the ~120 MB/s barrier, but several hover just below it. Several of the tests have very low transfer rates, and that's due to the nature of the test. The Content Creation test for example, simulates a user creating a web page, accessing multiple sources for the content. This kind of NAS device is built for handling in the neighborhood of 30 users doing content creation, so I'm a little suspicious about this test, but we'll see how it works with multiple scenarios before we rule it out. The Directory Copy tests use several hundred directories and several thousand files to test a typical backup and restore scenario.

QNAP_TS-879U-RP_Turbo_NAS_10GbE_Server_NASPT_1GbE_P15.png

Moving up the performance ladder, with the 10GbE connection in place, the results show significant gains across the board. The HD Video Playback tests all take full advantage of the larger Ethernet pipeline and posted improvements on the order of 4x, which is the same effect we saw in the earlier file transfer tests. Content Creation got a 2.5x boost, while the Directory Copy tests got a 2x improvement on copies from the NAS and performed 3.5x better on copies to the NAS. Clearly there are some significant factors in the test protocols that are emphasizing different aspects of NAS performance, otherwise all the results would be in the 400-500 MB/s range. As we generate more data with this benchmarking tool, we should start to see some performance patterns emerging.

QNAP_TS-879U-RP_Turbo_NAS_10GbE_Server_NASPT_10GbE_P16.png

Adding in AES-256 bit volume encryption to the equation, the peak test results fall back to levels more like those we saw with the GbE network connection. The more context-sensitive tests, like Content Creation and Directory Copy to the NAS all posted decent gains, though. On average, the numbers are about 50% better than the baseline performance with the GbE connection. Unfortunately, the average with encrypted drives is also only about half of what is possible without it. That's quite a performance hit, so be very sure to make an informed and rational decision before deciding to encrypt all your data. Most applications for this kind of NAS system are going to have a number of physical security safeguards in place that would negate the need for encryption. Most data centers are card-key access only, and have cameras pointed down every aisle, but all it takes is one very smart, rogue employee to make everyone wish that someone had taken the extra step to secure their data.

QNAP_TS-879U-RP_Turbo_NAS_10GbE_Server_NASPT_10GbE_AES_P17.png

To wrap things up, here's a consolidated chart of the "Fast" NASPT tests, and another for the "Slow" ones. First, the tests with relatively fast transfer rates. Once again, the performance of eight un-encrypted disks pumping data through a 10 Gigabit pipe is miles ahead of the alternatives. The larger bandwidth of 10GbE also gives the AES-256 encryption tests a small advantage over the baseline performance of the TS-879U-RP with its integral GbE connections. It's also interesting to see how the AES-256 performance goes up slightly as the video playback speed increases from 1x to 4x.

QNAP_TS-879U-RP_Turbo_NAS_10GbE_Server_NASPT_Fast_01.jpg

The "Slow" tests show a much narrower range of performance variation between the three different configurations. It also shows that most of these benchmarks got a more significant boost on the AES-256 scores than we saw in the "Fast" test results. The Directory Copy In results are a bit of an anomaly, but aren't completely out of the realm of possibilities. It does make me wonder if the metadata for the directories is encrypted. That might be a problem.....

QNAP_TS-879U-RP_Turbo_NAS_10GbE_Server_NASPT_Slow_01.jpg

Before we leave this benchmark, take a look at the network bandwidth chart from the QNAP Resource Monitor, as it records the data flows in and out of the NAS during the NASPT benchmark. The green trace is for data coming into the NAS, and the pink trace is for data leaving the NAS. Note the fifth peak, which represents the data for "HD Playback & Record" and you can see data simultaneously going into and out of the NAS. The big gap before any more data transfer is all the machinations going on in the Content Creation test - all that human simulation stuff that keeps the overall transfer rate low for this benchmark.

QNAP_TS-879U-RP_Turbo_NAS_10GbE_Server_NASPT_NIC_P20.png

This was my first real set of official tests with the Intel NASPT benchmarking tool, and I'm reasonably happy with the results. I'm not too pleased that two of the tests are so dependent on the amount of memory in the host PC, to the point of making their test results completely unusable. I'm also not willing to hobble the Windows 7 test bench to 2GB of RAM just to run this test suite. The fact that 10 of the 12 tests are not affected by this issue means I will choose to ignore the HD Video Record and File Copy to NAS results unless a patch is issued to fix this problem. All of the other tests give predictable and meaningful results that represent real-world scenarios, so I think I'll continue to use this benchmark in the future.

NAS Comparison Products



 

Comments 

 
# Ok test 10GbE but ..Federico La Morgia 2012-03-28 01:54
Test with 8xocz agility3 raid-0 for the maximum performance on the transfer rate 10GbE???
Report Comment
 
 
# RE: Ok test 10GbE but ..Olin Coles 2012-03-28 07:03
Sure! Would you like our address so you can send them to us?
Report Comment
 
 
# RE: RE: Ok test 10GbE but ..Federico La Morgia 2012-03-28 07:36
Unfortunately I have not, however, try to ask them directly to OCZ
Report Comment
 
 
# Might be fun, but...Bruce 2012-03-28 07:23
The results would be totally unrealistic. The great majority of users of this hardware are going to be stuffing it with mechanical HDDs, just because they need the capacity. I haven't seen any 2TB or 3TB SSDs around, have you? Also, anyone using an SSD in this type of application is going to have to use a very limited subset of SSDs - models that are specificallly designed for hard 24/7 RAID usage, without any TRIM support to keep the NAND cells fresh. The AGILITY uses "budget" flash memory, and any data center systems engineer who specified one for this kind of usage would be fired for incompetence.

So, I would be happy to use the QNAP TS-879U-RP to test some SSD makers' new enterprise-class drives, and run them hard, in a realistic test case. But, just stuffing some consumer grade devices in the NAS to push it closer to 10Gbps throughput doesn't really do much for me.... I "get" why QNAP tested it that way, but I also think it would have been useful for them to publish additional test results with enterprise-class HDDs.
Report Comment
 
 
# RE: Might be fun, but...Federico La Morgia 2012-03-28 07:40
the case of 8xSSD SATA3 in Raid-0 (as well as also the same in Raid-6 both normal and degraded discs 1-2) is used to understand the goodness of the disk controller present and / or the ultizzo cpu in operations in which need for high performance!
Use the SSD so you need to understand the physical limitations inherent in the product, it is obvious that no one ever use this product with SSDs, but the fact remains that with the HDD will never get to know the limits of the Qnap as well as any 'NAS or other product that has or needs to introduce SATA mass storage.
Report Comment
 
 
# SSD vertex 3Guy-Michel 2012-07-16 06:21
Hi we are planning use qnap either 879 or ts-ec879U-RP as a san unit for vmware it realistic? To boost performance we would could reuse ocz vertex 3 and 4 480.Second choice would be to put 480 ssd deneva from ocz for are database terminal server and accounting. would you recommand that? Or should we check to put the fastest sata available in raid 10 if supported.
Report Comment
 
 
# RE: SSD vertex 3Olin Coles 2012-07-16 06:54
Please provide a little more detail. Will you use only one SSD, or several in RAID? Will the NAS receive backups, or merely run with redundancy?
Report Comment
 
 
# detail in my mind :-)Guy-Michel 2012-07-16 07:09
would say that the main goal would be to have an array 4x480 vertex 4 let's say. the second array would be raid 10 fastest sata 6 we can find like Hitachi Deskstar 7K3000 or barracuda for file server and less critical vm machine.
Report Comment
 
 
# further more detailGuy-Michel 2012-07-16 07:28
Just to complete obove my plan is to use the qnap like SAN for datastore in vmware esxi5.If test is speed wise acceptable we gone build arround that.While re-using are actual vmhost with individual ssd drive. we gone put these ssd in a QNAP with ISCSI SAN probably upgrade the ethernet to 10gige adapter and finaly bring a cluster for HA.that the overall plan. Be able to acheive HA whitout breaking the bank.
Report Comment
 
 
# HA...??Bruce 2012-07-16 07:37
What is HA ?? I have never heard that acronym.
Report Comment
 
 
# RE: HA...??Guy-Michel 2012-07-16 07:39
it is a accronym (H)igh (A)vailability.
Report Comment
 
 
# RE: detail in my mind :-)Bruce 2012-07-16 07:32
Even though it says on the QNAP website: "...QNAP NAS supports advanced RAID configurations and multiple RAID volumes on a single NAS." I don't believe you can actually set up two separate RAID arrays on one NAS. Is that what you are proposing, two arrays of four drives each one one TS-879 machine? I don't think that's possible. Try posting that question on the QNAP forums, one of the QNAP representatives will give you confirmation.

You CAN set up multiple iSCSI targets and/or multiple LUNs one one device, but they would all reside on one physical RAID volume.
Report Comment
 
 
# Lucky You...Bruce 2012-07-16 07:00
I would use the Enterprise Class SSDs if you can afford it. They have superior wear-leveling routines built into them. Remember, the operating system on the QNAP is not doing anything to keep the NAND refreshed on the SSDs, unlike consumer systems. I know OCZ and other vendors have been improving the "wear resistance" of all their SSD offerings, but the OCZ Denaeva will still have the most capable systems for keeping the performance up in this sort of usage.

RAID 10 (0r RAID 20)is usually best for database applications. RAID 5 can be slower in Write operations. Do you have the ability to set the system up in a test environment? I would strongly encourage that, so you can try the different configurations.
Report Comment
 
 
# QNAP 10GbE PromotionBruce 2012-04-06 07:12
BTW, QNAP is bundling a 10GbE NIC with some models.
Details here: #qnap.com/static/landing/10gbe_en.html

Did I inspire them...? LOL
Report Comment
 
 
# RE: QNAP 10GbE PromotionSébastien 2012-06-08 02:04
Any idea what transfer rate could be achieved if USB 3.0 was used instead of 10 Gbe (RAID5, same disks)?
Report Comment
 
 
# At the max, maybeBruce 2012-06-08 08:10
From Wikipedia: The "SuperSpeed" bus provides a transfer mode at 5.0 Gbit/s additionally to the three existing transfer modes. The raw throughput is 4 Gbit/s, and the specification considers it reasonable to achieve 3.2 Gbit/s (0.4 GB/s or 400 MB/s) or more.

I got more than 450MB/s in ACTUAL real-world throughput, which is slightly more than the USB-IF expects the USB 3.0 connection to handle, so I would say that using USB instead of Ethernet would throttle the bandwidth somewhat. Of course, you lose all the advantages of having the device sitting directly on the network, which is a major feature of this and any other NAS.
Report Comment
 
 
# USB 3.0Sébastien 2012-06-08 10:22
You are right but my plan is to share the NAS between 5 computers with slow streaming capability (connected to a 1 Gbe switch) and one or two close workstations with fast streaming capability (connected to the USB 3.0 ports). Do you know if this hybrid mode is supported? Do you confirm 450 MB/s with the NAS configuration you described and USB 3.0?

In case of simultaneous streaming what total throughput can I expect... could the NAS handle 2x450 + 100 MB/s = 1000 MB/s? That should be supported by 8 high end disks but I do not know if the processor can handle a RAID5 encoding/decoding at this rate...

Last question: is it possible to wire the NAS with two 1 Gbe cables to the switch and handle two 100 MB/s streams from two different computers? Is it seamless - I mean would the computers all see a single disk or is it more complex to aggregate?
Report Comment
 
 
# Not recommendedBruce 2012-06-08 11:48
You are not going to be able to manage all these data streams with a quasi-network of USB connections connected to the TS-879U-RP.

Your best bet would be to get a 10GbE switch, like the one I mentioned from Cisco, in the review. That way, all your workstations can get the bandwidth they need, and you have the bandwidth for future expansion.
Report Comment
 
 
# USB 3.0Sébastien 2012-06-08 12:35
Five of the computers connected to the NAS are old and run LOW profile hard drives (max 50 MB/s), it is very unlikely that more than 2 of them access the NAS simultaneously and none is equipped with 10Gbe adapters. Only a single (possibly 2 in the future) workstation is able to transfer data at a rate over 500 MB/s and it is the computer I expect to transfer the largest amount of data.

Given this situation I think that the 10Gbe switch + many 10Gbe adapters is overkilling... I thought USB 3.0 would be very well suited to this kind of unsymmetric and non simulataneous access scenario. Where do you exactly see a problem? Do you think that the hybrid mode cannot work in practice or were you only saying the NAS cannot handle two USB 3.0 streams at 450 MB/s?
Report Comment
 
 
# Not too many 10GbE....Bruce 2012-06-08 13:02
The 10GbE switch I mentioned has only two 10GbE ports, the rest are all just GbE. That way, the individual workstation loads get aggreagated by the switch. The 10GbE ports are usually just fot rhe NAS, but you could put your workstation into one of them.

My main issue is that all the software for this, and most NAS devices, is designed and optimized to work in an Ethernet environment, not an ad-hoc USB network. The capabilities while connected via USB will be severly restricted. Upload and download, via a couple of built-in scripts, that's all.

FWIW, I'm never been impressed by the performance of USB with external drives. I guess I should test this one before I pull it apart...
Report Comment
 
 
# RE: Not too many 10GbE....Sébastien 2012-06-08 13:38
I get your point and I have no doubt 10Gbe is well more efficient than USB 3.0 but I guess the price of the 10Gbe switch you mention would be well over the price of the NAS itself so I am afraid I cannot afford this extra cost. Still I would be really happy with the 450 MB/s you reported over USB 3.0 and I guess it will not interfere with the traffic over the 1Gbe port since the transfer will (a priori) not be simultaneous. My worry is more about the configuration of the system in ethernet / USB hybrid mode: is it prepared for this or is it thought to work exclusively in USB or ethernet mode?

Note: If you ever get the chance to test dual USB 3.0 transfer from two different computers ... :)

Thanks a lot for your help!
Report Comment
 
 
# IT ConsultantsHenri 2012-11-14 23:26
Hi,
have such a box installed as ESXI 5.1 storage. Monday this week, the ESXI 5.1 freezed about 15 times after installing a additional RAID0 with 2 Samsung 830 512GB SSDs. Box is attached by 2* 10Gbit Intel X520 DA2.
Stopped to access this DataStore seams to fix the problem.
Support of QNAP always recommands to reset the box to the factory setup. Not very helpful.
Report Comment
 
 
# Multiple RAID VolumesBruce Normann 2012-11-15 06:35
Does the firmware you are using support multiple RAID volumes? Last time I checked you couldn't do that. Multiple LUNs, yes....but that's not the same thing as a new RAID volume. There are some devices out there thad support multiple volumes, just not sure if QNAP added that capability since I last inquired.
Report Comment
 

Comments have been disabled by the administrator.

Search Benchmark Reviews
QNAP Network Storage Servers

Follow Benchmark Reviews on FacebookReceive Tweets from Benchmark Reviews on Twitter