Original Link: https://www.anandtech.com/show/8799/qnap-ts853-pro-8bay-intel-bay-trail-smb-nas-review
QNAP TS-853 Pro 8-bay Intel Bay Trail SMB NAS Review
by Ganesh T S on December 29, 2014 7:30 AM EST
Introduction and Testbed Setup
QNAP has focused on Intel's Bay Trail platform for this generation of NAS units (compared to Synology's efforts with Intel Rangeley). While the choice made sense for the home users / prosumer-targeted TS-x51 series, we were a bit surprised to see the TS-x53 Pro series (targeting business users) also use the same Bay Trail platform. Having evaluated 8-bay solutions from Synology (the DS1815+) and Asustor (the AS7008T), we requested QNAP to send over their 8-bay solution, the TS-853 Pro-8G. Hardware-wise, the main difference between the three units lie in the host processor and the amount of RAM.
The specifications of our sample of the QNAP TS-853 Pro are provided in the table below
| QNAP TS-853 Pro-8G Specifications | |
| Processor | Intel Celeron J1900 (4C/4T Silvermont x86 @ 2.0 GHz) |
| RAM | 8 GB |
| Drive Bays | 8x 3.5"/2.5" SATA II / III HDD / SSD (Hot-Swappable) |
| Network Links | 4x 1 GbE |
| External I/O Peripherals | 3x USB 3.0, 2x USB 2.0 |
| Expansion Slots | None |
| VGA / Display Out | HDMI (with HD Audio Bitstreaming) |
| Full Specifications Link | QNAP TS-853 Pro-8G Specifications |
| Price | USD 1195 |
Note that the $1195 price point is for the 8GB RAM version. The default 2 GB version retails for $986. The extra RAM is important if the end user wishes to take advantage of the unit as a VM host using the Virtualization Station package.
The TS-853 Pro runs Linux (kernel version 3.12.6). Other aspects of the platform can be gleaned by accessing the unit over SSH.
Compared to the TS-451, we find that the host CPU is now a quad-core Celeron (J1900) instead of a dual-core one (J1800). The amount of RAM is doubled. However, the platform and setup impressions are otherwise similar to the TS-451. Hence, we won't go into those details in our review.
One of the main limitations of the TS-x51 units is the fact that it can have only one virtual machine (VM) active at a time. The TS-x53 Pro relaxes that restriction and allows two simultaneous VMs. Between our review of the TS-x51 and this piece, QNAP introduced QvPC, a unique way to use the display output from the TS-x51 and TS-x53 Pro series. We will first take a look at the technology and how it shaped our evaluation strategy.
Beyond QvPC, we follow our standard NAS evaluation routine - benchmark numbers for both single and multi-client scenarios across a number of different client platforms as well as access protocols. We have a separate section devoted to the performance of the NAS with encrypted shared folders, as well as RAID operation parameters (rebuild durations and power consumption). Prior to all that, we will take a look at our testbed setup and testing methodology.
Testbed Setup and Testing Methodology
The QNAP TS-853 Pro can take up to 8 drives. Users can opt for either JBOD, RAID 0, RAID 1, RAID 5, RAID 6 or RAID 10 configurations. We expect typical usage to be with multiple volumes in a RAID-5 or RAID-6 disk group. However, to keep things consistent across different NAS units, we benchmarked a single RAID-5 volume across all disks. Eight Western Digital WD4000FYYZ RE drives were used as the test disks. Our testbed configuration is outlined below.
| AnandTech NAS Testbed Configuration | |
| Motherboard | Asus Z9PE-D8 WS Dual LGA2011 SSI-EEB |
| CPU | 2 x Intel Xeon E5-2630L |
| Coolers | 2 x Dynatron R17 |
| Memory | G.Skill RipjawsZ F3-12800CL10Q2-64GBZL (8x8GB) CAS 10-10-10-30 |
| OS Drive | OCZ Technology Vertex 4 128GB |
| Secondary Drive | OCZ Technology Vertex 4 128GB |
| Tertiary Drive | OCZ Z-Drive R4 CM88 (1.6TB PCIe SSD) |
| Other Drives | 12 x OCZ Technology Vertex 4 64GB (Offline in the Host OS) |
| Network Cards | 6 x Intel ESA I-340 Quad-GbE Port Network Adapter |
| Chassis | SilverStoneTek Raven RV03 |
| PSU | SilverStoneTek Strider Plus Gold Evolution 850W |
| OS | Windows Server 2008 R2 |
| Network Switch | Netgear ProSafe GSM7352S-200 |
The above testbed runs 25 Windows 7 VMs simultaneously, each with a dedicated 1 Gbps network interface. This simulates a real-life workload of up to 25 clients for the NAS being evaluated. All the VMs connect to the network switch to which the NAS is also connected (with link aggregation, as applicable). The VMs generate the NAS traffic for performance evaluation.
Thank You!
We thank the following companies for helping us out with our NAS testbed:
- Thanks to Intel for the Xeon E5-2630L CPUs and the ESA I-340 quad port network adapters
- Thanks to Asus for the Z9PE-D8 WS dual LGA 2011 workstation motherboard
- Thanks to Dynatron for the R17 coolers
- Thanks to G.Skill for the RipjawsZ 64GB DDR3 DRAM kit
- Thanks to OCZ Technology for the two 128GB Vertex 4 SSDs, twelve 64GB Vertex 4 SSDs and the OCZ Z-Drive R4 CM88
- Thanks to SilverStone for the Raven RV03 chassis and the 850W Strider Gold Evolution PSU
- Thanks to Netgear for the ProSafe GSM7352S-200 L3 48-port Gigabit Switch with 10 GbE capabilities.
- Thanks to Western Digital for the eight WD RE hard drives (WD4000FYYZ) to use in the NAS under test.
QNAP's HD Station - QvPC Explored
Over the last few years, QNAP has been introducing various interesting packages for their QTS operating system. XBMC was one of the first to take advantage of the HDMI display output on their NAS units. Slowly, a few other packages such as web browsers, NAS management interfaces and YouTube were introduced. Recently, QNAP decided to bundle all these packages under the Hybrid Desk Station (HD Station) moniker.
The ability to interact with the NAS using a keyboard and a mouse brought it closer to a traditional PC for the average consumer. Virtualization Station was the final piece in the puzzle. QNAP added an option to display the console of each VM on the display output desktop (a QVM option) in the Virtualization Station settings. This enables end users to interact not only with the QTS desktop, but also with a Windows or Ubuntu VM as if it was running on a physical machine.
QNAP's HD Station Interface with QVM Activated
Recognizing the big leap made in HD Station using QVMs, QNAP has opted to do a branding exercise - QvPC - for this feature set. In addition to QVMs, HD Station also includes a local output for the tracked surveillance cameras, XBMC with HD audio bitstreaming and other miscellaneous features. QNAP's product page for HD Station adds more marketing detail to what we have covered above.
Windows 10 on a QNAP TS-853 Pro-8G - QvPC
Our usual NAS evaluation routine doesn't load up the NAS with non-NAS aspects while processing benchmarks. For the TS-853 Pro-8G, we did one pass with all four network links bonded and no extra packages active. However, with running VMs being a very attractive feature of the TS-x53 Pro series, we added two sets of benchmarks for the multi-client scenarios after configuring two Windows 10 VMs (two cores, 2 GB RAM and one dedicated network link each). This configuration left the other two network ports free for bonding / external traffic. The CPU cores themselves had to be shared between the VMs and the main NAS tasks (as all four cores ended up getting allocated to the VMs).
- 1 QVM: A Windows 10 VM with QVM configured was activated. PC Mark 8's 'Work' workload was run in a loop with HDMI output enabled, while the rest of the NAS was subject to storage traffic over the bonded network links
- 2 VMs: QVM was de-activated. Two Windows 10 VMs were configured for access via Remote Desktop. PC Mark 8's 'Work' workload was run in a loop on both VMs, while the rest of the NAS was subject to storage traffice over the bonded network links.
In addition to the above Virtualization Station configuration, we also found an interesting caching aspect in QTS. Our benchmarks are processed after turning off the write cache (EXT4 delay allocation) for all NAS units. However, QTS, in addition to the EXT4 delay allocation, has an additional system cache setting at the volume level. The large amount of RAM in the system (8 GB) also skewed the results quite a bit with the system cache settings enabled. QNAP is also one of the first COTS NAS vendors to support SMB 3.0. Obviously, this is available only for Windows 8 and Windows Server 2012 R2 and later operating systems. We tested out SMB 3.0 with a Windows 8 VM in our NAS testbed. In the rest of the review, we try to address as many different combinations as possible in the benchmarks.
Single Client Performance - CIFS & iSCSI on Windows
The single client CIFS and iSCSI performance of the QNAP TS-853 Pro 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. SMB 3.0 performs very well, but, with the host OS being Windows 8, it wouldn't really be fair to compare it against other NAS units that were processed with a Windows 7 client. In any case, we find enabling SMB 3.0 encryption pulls down the processing rate to around 20 MBps irrespective of the type of traffic. Enabling VMs pulls down the performance. In general, Haswell performs better than Rangeley or Bay Trail, but Asustor's ADM is not yet optimized fully. This allows Synology and QNAP to pull ahead with their Rangeley / Bay Trail solutions.
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 observations we had in the CIFS subsection above hold true here too.
iSCSI testing results closely track the CIFS test results. Note that we only performed evaluation with Windows 7. The system cache was turned on for these tests (though the EXT4 delay allocation was disabled).
Single Client Performance - CIFS and NFS on Linux
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 -aczR -g 2097152 -U /PATH_TO_LOCAL_CIFS_MOUNT -f /PATH_TO_LOCAL_CIFS_MOUNT/testfile -b <NAS_NAME>_CIFS_EXCEL_BIN.xls > <NAS_NAME>_CIFS_CSV.csv
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-853 Pro - Linux Client Performance (MBps) | ||
| IOZone Test | CIFS | NFS |
| Init Write | 78 | 79 |
| Re-Write | 83 | 81 |
| Read | 45 | 115 |
| Re-Read | 48 | 114 |
| Random Read | 29 | 64 |
| Random Write | 81 | 78 |
| Backward Read | 28 | 51 |
| Record Re-Write | 1648* | 1653* |
| Stride Read | 44 | 108 |
| File Write | 83 | 80 |
| File Re-Write | 82 | 81 |
| File Read | 32 | 93 |
| File Re-Read | 33 | 94 |
| *: Benchmark number skewed due to caching effect | ||
Multi-Client Performance - CIFS on Windows
We put the QNAP TS-853 Pro through some IOMeter tests with a CIFS share being accessed from up to 25 VMs simultaneously. The following four graphs show the total available bandwidth and the average response time while being subject to different types of workloads through IOMeter. The tool also reports various other metrics of interest such as maximum response time, read and write IOPS, separate read and write bandwidth figures etc. Detailed listings of the IOMeter benchmark numbers (including IOPS and maximum response times) for each configuration are linked below:
- QNAP TS-853 Pro - 8 HDDs / RAID-5 - 4x 1G
- QNAP TS-853 Pro - 8 HDDs / RAID-5 - 2 VMs / 2x 1G
- QNAP TS-853 Pro - 8 HDDs / RAID-5 - 1 QVM / 2x 1G
The important aspect to note here is that the performance for the random workloads when the VMs are active are all over the place, following no particular pattern. This is due to the fact that the PC Mark 8 'Work' workload running in the background doesn't uniformly load the system resources. It is sufficient to observe that even moderately heavy word processing or similar task can pull down the NAS performance for certain types of workloads.
In the absence of active VMs, enabling link aggregation with all four ports allows maximum throughput numbers in the order of 400+ MBps for pure read workloads.
Multi-Client iSCSI Evaluation
As virtualization becomes more and more popular even in home / power user settings, the importance of the iSCSI feature set of any COTS NAS can't be overstated. Starting with our ioSafe 1513+ review, we have started devoting a separate section (in the reviews of NAS units targeting SMBs and SMEs) to the evaluation of iSCSI performance. QNAP supports only one type of iSCSI LUN - regular files.
We evaluated the performance of the QNAP TS-853 Pro with file-based LUNs. The standard IOMeter benchmarks that we used for multi-client CIFS evaluation were utilized for iSCSI evaluation also. The main difference to note is that the CIFS evaluation was performed on a mounted network share, while the iSCSI evaluation was done on a 'clean physical disk' (from the viewpoint of the virtual machine).
Performance Numbers
The four IOMeter traces were run on the physical disk manifested by mapping the iSCSI target on each VM. The benchmarking started with one VM accessing the NAS. The number of VMs simultaneously playing out the trace was incremented one by one till we had all 25 VMs in the fray. Detailed listings of the IOMeter benchmark numbers (including IOPS and maximum response times) for each configuration are linked below:
QNAP's implementation provides very good results for purely sequential workloads compared to Synology. However, the latter wins out when it comes to random workloads. On the whole, the performance of the iSCSI LUNs in QTS is satisfactory. Features such as VAAI and ODX are available. However, Synology manages to edge QNAP out when it comes to offering multiple ways of configuring iSCSI LUNs. These ways provide different balancing acts for performance and flexibility.
Encryption Support Evaluation
Consumers looking for encryption capabilities can opt to encrypt a iSCSI share with TrueCrypt or some in-built encryption mechanism in the client OS. However, if requirements dictate that the data must be shared across multiple users / computers, relying on encryption in the NAS is the best way to move forward. Most NAS vendors use the industry-standard 256-bit AES encryption algorithm. One approach is to encrypt only a particular shared folder while the other approach is to encrypt the full volume. QNAP suports only volume-level encryption.
On the hardware side, encryption support can be in the form of specialized hardware blocks in the SoC (common in ARM / PowerPC based NAS units). In x86-based systems, accelerated encryption support is dependent on whether the AES-NI instruction is available on the host CPU. Bay Trail / Celeron J1900 doesn't have AES-NI capabilities. Therefore, encrypted volume processing in the TS-853 Pro is done purely via multi-threaded software.
Spreading the encryption / decryption operations across four cores manages to help QNAP avoid much of a penalty when compared to the performance of a non-encrypted volume. That said, performance is bound to show marked decline with active VMs and/or simultaneous multi-client accesses.
Miscellaneous Aspects and Final Words
In order to keep testing consistent across all 8-bay units, we performed all our expansion / rebuild testing as well as power consumption evaluation with the unit configured in RAID-5. The disks used for benchmarking (Western Digital WD4000FYYZ) were also used in this section. The table below presents the average power consumption of the unit as well as time taken for various RAID-related activities.
| QNAP TS-853 Pro RAID Expansion and Rebuild / Power Consumption | ||
| Activity | Duration (HH:MM:SS) | Avg. Power (W) |
| Single Disk Init | 00:02:51 | 30.34 W |
| JBOD to RAID-1 Migration | 08:32:42 | 39.72 W |
| RAID-1 (2D) to RAID-5 (3D) Migration | 23:07:23 | 50.41 W |
| RAID-5 (3D) to RAID-5 (4D) Expansion | 01:00:29 | 59.69 W |
| RAID-5 (4D) to RAID-5 (5D) Expansion | 01:50:43 | 71.01 W |
| RAID-5 (5D) to RAID-5 (6D) Expansion | 01:57:20 | 81.57 W |
| RAID-5 (6D) to RAID-5 (7D) Expansion | 02:08:18 | 91.9 W |
| RAID-5 (7D) to RAID-5 (8D) Expansion | 01:59:59 | 101.88 W |
| RAID-5 (8D) Rebuild | 10:56:15 | 101.62 W |
The graphs below show the power consumption and rebuild duration when repairing a RAID-5 volume for the various 8-bay NAS units that have been evaluated before.
QNAP's RAID rebuild times have consistently been great (even with the bitmap feature turned off). We see that the duration is beat only by Asustor's AS7008T. However, power consumption is quite a bit lower for the TS-853 Pro. In fact, the TS-853 Pro is the most power efficient of the lot when it comes to RAID-5 rebuild.
Concluding Remarks
Coming to the business end of the review, we are in a good position to discuss the merits of various 8-bay solutions for the mid-range SMB market. The main tussle is between the Synology DS1815+ and the QNAP TS-853 Pro. When it comes to home and power users / low-end SMBs, the TS-853 Pro can be easily recommended. The real-time transcoding and media serving capabilities as well as the QNAP suite of mobile apps get the job done without much ado. Virtualization Station is an awesome differentiating feature - something that the competition is yet to match. Applications for VMs are numerous - home automation controllers, casual software development machines - the list goes on. We also tested out system migration by moving disks from our TS-451 sample under long term testing to the TS-853 Pro, and the new system was up and running with all data / user details intact immediately.
However, when it comes to the actual target market - mid-range SMBs, it looks like the TS-853 Pro has missed the mark with the current version of QTS. One of the most glaring shortcomings is the absence of the Storage Pool feature - something available in the higher end units such as the TS-EC1279U-RP. The other aspect is the limited iSCSI LUN configuration capabilities. Synology really hits a home run with their iSCSI feature set. Fortunately, QNAP indicated that both of these aspects are being addressed in QTS 4.2, slated to go public in the next few months.
The final verdict is that if you are a home / SOHO user looking for a powerful NAS solution, the TS-853 Pro should be your go-to solution if you can afford a mid-range SMB price. Business users should wait for QTS 4.2 to become public before making a firm decision in favor of the TS-853 Pro. [ UPDATE: QNAP's QTS 4.1.2 update brings block-based LUNs as well as storage pools support to the TS-853 Pro ].
Facebook
Twitter
RSS
