Introduction and Testbed Setup

Western Digital is a comparatively recent entrant to the fast-growing NAS market. Despite having had a Windows Storage Server-based product for business users for some time now, a custom embedded-Linux based SMB-targeted model was lacking. Avoiding Windows in the NAS helps bring down the cost of the unit and also makes the units easier to manage for small businesses without dedicated IT staff. Last week, they officially launched the My Cloud EX2100 (2-bay) and EX4100 (4-bay) NAS units to target this market.

The chassis design of the EX4100 is very similar to that of the EX4 that we reviewed last year. Despite a smaller height and larger width, the design of the drive bays and the I/O ports are essentially the same. There is no drive caddy (which means that only 3.5" drives are supported). There is a information screen in the front panel with up and down buttons to navigate the current status messages. The main difference between the EX4 and the DL4100 is the presence of a USB 3.0 port along with a one-touch copy button in the front panel. The gallery below takes us around the contents of the package and the chassis design. Our review unit came with 6 TB WD Red drives pre-installed and pre-configured. The unit uses a 90W (19V @ 4.74A) power adapter.

The specifications of the WD My Cloud DL4100 are provided in the table below

WD My Cloud DL4100 Specifications
Processor	Intel Atom C2338 (2C/2T Silvermont x86 @ 1.7 GHz)
RAM	2 GB
Drive Bays	4x 3.5" SATA II / III HDD / SSD (Hot-Swappable)
Network Links	2x 1 GbE
External I/O Peripherals	2x USB 3.0, 1x USB 2.0
Expansion Slots	N/A
VGA / Display Out	N/A
Full Specifications Link	WD My Cloud DL4100 Specifications
Price	USD 530

Note that the $530 pricing is for the diskless unit. The 8 TB version sells for $850, 16 TB for $1170 and 24 TB for $1529.

Western Digital provides power users with SSH access to the unit, and this gives us some more insight into the platform.

WD uses Linux kernel version 3.10.38 in their 64-bit OS build. Even though the unit comes only with 2 GB of RAM, users can install a DDR3L SO-DIMM in one of the empty slots to push it up to 6 GB of RAM. Since the Rangeley SoC (Atom C2338) doesn't have any integrated USB 3.0 ports, the board must definitely be sporting a PCIe - USB 3.0 bridge. The SoC has support for up to 4x 1GbE ports, but does need external PHYs. A pair of Marvell Alaska 88E1512 PHYs are on board for this purpose. A 2 GB Micron SLC NAND flash chip holds the OS of the unit.

In the rest of the review, we will take a look at the 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. Prior to all that, we will take a look at our testbed setup and testing methodology.

Testbed Setup and Testing Methodology

The WD My Cloud DL4100 can take up to 4 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 a RAID-5 or RAID-6 volume. However, to keep things consistent across different NAS units, we benchmarked a RAID-5 volume (i.e, single disk redundancy mode). Four Western Digital WD4000FYYZ RE drives were used as the test disks, even though our review unit shipped with 6 TB WD Red drives. 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.

Single Client Performance - CIFS & iSCSI on Windows

The single client CIFS and iSCSI performance of the WD My Cloud DL4100 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. One of the interesting aspects of the My Cloud OS is the support for SMB 3.0. Amongst the NAS units that we have evaluated, this is the only OS with support for the latest SMB version other than QNAP's QTS 4.x. SMB 3.0 was evaluated using a Windows 8 VM. It obviously provides better performance compared to SMB 2.0 (Windows 7 doesn't support SMB 3.0).

If the SMB 3.0 results are ignored, we find that the WD My Cloud DL4100 is consistently bettered by either the Synology DS415+ (based on a faster Rangeley SoC) or the QNAP TS-451 (based on a higher-clocked Bay Trail Celeron). In addition, Synology's DSM and QNAP's QTS are much more mature compared to the My Cloud OS.

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.

The iSCSI implementation in the My Cloud OS has lot of scope for improvement. In certain workloads, even the ARM-based LenovoEMC ix4-300d manages to score better numbers. Other than that, the behavior of the performance numbers track what was observed for the CIFS benchmarks.

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.

WD My Cloud DL4100 - Linux Client Performance (MBps)
IOZone Test	CIFS	NFS
Init Write	78	35
Re-Write	74	38
Read	43	116
Re-Read	42	117
Random Read	25	65
Random Write	66	29
Backward Read	26	52
Record Re-Write	1547*	256*
Stride Read	40	113
File Write	74	38
File Re-Write	74	39
File Read	30	95
File Re-Read	30	96
*: Benchmark number skewed due to caching effect

Multi-Client Performance - CIFS on Windows

We put the WD My Cloud DL4100 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:

• WD My Cloud DL4100 - 4 HDDs / RAID-5 / 2x 1G

WD My Cloud DL4100 - 2x 1G - 100% Sequential ReadsSeagate NAS 4-bay - 2x 1G - 100% Sequential ReadsSeagate NAS Pro 4-bay - 2x 1G - 100% Sequential ReadsAsustor AS-304T - 1x 1G - 100% Sequential ReadsThecus N4800 - 2x 1G - 100% Sequential ReadsQNAP TS-451 - 2x 1G - 100% Sequential ReadsLenovoEMC ix4-300d - 2x 1G - 100% Sequential ReadsSynology DS414j - 1x 1G - 100% Sequential ReadsWD My Cloud EX4 - 2x 1G - 100% Sequential ReadsSynology DS415+ - 2x 1G - 100% Sequential Reads

 

WD My Cloud DL4100 - 2x 1G - Max Throughput - 50% ReadsSeagate NAS 4-bay - 2x 1G - Max Throughput - 50% ReadsSeagate NAS Pro 4-bay - 2x 1G - Max Throughput - 50% ReadsAsustor AS-304T - 1x 1G - Max Throughput - 50% ReadsThecus N4800 - 2x 1G - Max Throughput - 50% ReadsQNAP TS-451 - 2x 1G - Max Throughput - 50% ReadsLenovoEMC ix4-300d - 2x 1G - Max Throughput - 50% ReadsSynology DS414j - 1x 1G - Max Throughput - 50% ReadsWD My Cloud EX4 - 2x 1G - Max Throughput - 50% ReadsSynology DS415+ - 2x 1G - Max Throughput - 50% Reads

 

WD My Cloud DL4100 - 2x 1G - Random 8K - 70% ReadsSeagate NAS 4-bay - 2x 1G - Random 8K - 70% ReadsSeagate NAS Pro 4-bay - 2x 1G - Random 8K - 70% ReadsAsustor AS-304T - 1x 1G - Random 8K - 70% ReadsThecus N4800 - 2x 1G - Random 8K - 70% ReadsQNAP TS-451 - 2x 1G - Random 8K - 70% ReadsLenovoEMC ix4-300d - 2x 1G - Random 8K - 70% ReadsSynology DS414j - 1x 1G - Random 8K - 70% ReadsWD My Cloud EX4 - 2x 1G - Random 8K - 70% ReadsSynology DS415+ - 2x 1G - Random 8K - 70% Reads

 

WD My Cloud DL4100 - 2x 1G - Real Life - 65% ReadsSeagate NAS 4-bay - 2x 1G - Real Life - 65% ReadsSeagate NAS Pro 4-bay - 2x 1G - Real Life - 65% ReadsAsustor AS-304T - 1x 1G - Real Life - 65% ReadsThecus N4800 - 2x 1G - Real Life - 65% ReadsQNAP TS-451 - 2x 1G - Real Life - 65% ReadsLenovoEMC ix4-300d - 2x 1G - Real Life - 65% ReadsSynology DS414j - 1x 1G - Real Life - 65% ReadsWD My Cloud EX4 - 2x 1G - Real Life - 65% ReadsSynology DS415+ - 2x 1G - Real Life - 65% Reads

The takeaway from these tests is that beyond 15 or so simultaneous clients, the My Cloud DL4100 doesn't exhibit great performance consistency. Compared to other NAS units, there are weaknesses in the sequential access patterns, but the numbers for the random workloads are better.

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. NAS vendors have multiple options when it comes to implementing iSCSI LUNs. By far, the most common and simplest method is in the form of regular files. Western Digital adopts that method too. WD also allows multiple initiators to connect to a single iSCSI target.

We evaluated the performance of the WD My Cloud DL4100 with file-based LUNs. In the first case, we configured a single LUN and had multiple initiators from different machines connect to it. In the second case, we created one LUN/target for each initiator to connect to. 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:

• WD My Cloud DL4100 - 4 HDDs / RAID-5 / 2x 1G - Single LUN (Regular File)
• WD My Cloud DL4100 - 4 HDDs / RAID-5 / 2x 1G - LUNs (Regular Files)

WD My Cloud DL4100 - Single LUN (Regular File) - 100% Sequential ReadsWD My Cloud DL4100 - LUNs (Regular Files) - 100% Sequential ReadsSeagate NAS Pro 4-bay - Single LUN (Regular File) - 100% Sequential ReadsSeagate NAS Pro 4-bay - LUNs (Regular Files) - 100% Sequential ReadsSynology DS415+ - LUNs (Regular Files) - 100% Sequential ReadsSynology DS415+ - Single LUN on RAID - 100% Sequential Reads

 

WD My Cloud DL4100 - Single LUN (Regular File) - Max Throughput - 50% ReadsWD My Cloud DL4100 - LUNs (Regular Files) - Max Throughput - 50% ReadsSeagate NAS Pro 4-bay - Single LUN (Regular File) - Max Throughput - 50% ReadsSeagate NAS Pro 4-bay - LUNs (Regular Files) - Max Throughput - 50% ReadsSynology DS415+ - LUNs (Regular Files) - Max Throughput - 50% ReadsSynology DS415+ - Single LUN on RAID - Max Throughput - 50% Reads

 

WD My Cloud DL4100 - Single LUN (Regular File) - Random 8K - 70% ReadsWD My Cloud DL4100 - LUNs (Regular Files) - Random 8K - 70% ReadsSeagate NAS Pro 4-bay - Single LUN (Regular File) - Random 8K - 70% ReadsSeagate NAS Pro 4-bay - LUNs (Regular Files) - Random 8K - 70% ReadsSynology DS415+ - LUNs (Regular Files) - Random 8K - 70% ReadsSynology DS415+ - Single LUN on RAID - Random 8K - 70% Reads

 

WD My Cloud DL4100 - Single LUN (Regular File) - Real Life - 65% ReadsWD My Cloud DL4100 - LUNs (Regular Files) - Real Life - 65% ReadsSeagate NAS Pro 4-bay - Single LUN (Regular File) - Real Life - 65% ReadsSeagate NAS Pro 4-bay - LUNs (Regular Files) - Real Life - 65% ReadsSynology DS415+ - LUNs (Regular Files) - Real Life - 65% ReadsSynology DS415+ - Single LUN on RAID - Real Life - 65% Reads

The performance issues we encountered in the single-client iSCSI benchmarks are evident here also. The other 4-bay NAS units we have put through these tests (the Seagate NAS Pro 4-bay as well as the Synology DS415+) exhibit better numbers as well as consistency compared to the WD My Cloud DL4100.

Towards the end of the iSCSI testing process, we ran a script on the clients' side to disconnect the iSCSI targets one by one, while refreshing the My Cloud DL4100's web UI's iSCSI page at the same time. During this process, the unit froze up completely - the web UI wasn't reachable, the LCD navigation froze and a SSH connection was refused - there was no way out but to yank out the power and restart the unit. Unfortunately, the logs reported only a power failure and this was not a repeatable issue. In addition to general performance improvements, it appears that the My Cloud OS could do with stability QA also.

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. Western Digital supports encryption on a volume basis only.

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. Rangeley SoCs have AES-NI support and the performance difference between the encrypted and non-encrypted volumes can be expected to be reasonable.

Read transactions don't have much penalty, but writes to encrypted volumes are much slower compared to non-encrypted volumes. For example, there is a 20 MBps difference in our standard robocopy test of writing a Blu-ray folder to the NAS. Compared to other NAS units, the raw numbers are very good, thanks to the hardware acceleration available. If other Rangeley-based units such as the Synology DS415+ or the Seagate NAS Pro 4-bay are considered, the numbers are more or less the same.

In our first pass with the NASPT benchmark, we saw that the mapped CIFS share from the encrypted volume randomly disconnected from the client during the middle of the fourth pass (the NASPT batch run has five passes of each test). Unfortunately,  this was again not a repeatable issue, as our second batch run completed without problems.

Miscellaneous Aspects and Final Words

In order to keep testing consistent across all 4-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.

WD My Cloud DL4100 RAID Expansion and Rebuild / Power Consumption
Activity	Duration (HH:MM:SS)	Avg. Power (W)
Single Disk Init	00:09:06	22.63 W
JBOD to RAID-1 Migration	08:37:27	35.34 W
RAID-1 (2D) to RAID-5 (3D) Migration	04:43:33	45.31 W
RAID-5 (3D) to RAID-5 (4D) Expansion	18:01:04	55.1 W
RAID-5 (4D) Rebuild	08:50:33	57.41 W

The graphs below show the power consumption and rebuild duration when repairing a RAID-5 volume for the various 4-bay NAS units that have been evaluated before.

Even though the power consumption during the rebuild process is high, the time taken is the lowest amongst all the 4-bay NAS units that we have seen so far. That said, the time take for expanding a 3-disk RAID-5 volume to a 4-disk RAID-5 volume is more than double the RAID-5 rebuild time. Usually, we see NAS units have approximately same duration and power consumption for these two activities. The My Cloud OS does have some scope for optimizing the expansion process.

Concluding Remarks

The My Cloud DL4100 fills an important slot in Western Digital's NAS lineup for the SMB / SOHO market. The choice of an Intel Rangeley platform enables very good performance numbers while keeping power consumption numbers reasonable. On the OS side, getting the unit up and running with relay access to the unit / mobile apps support was quite easy.

On the other side, the user experience with the My Cloud OS needs a lot of improvement. For example, navigating the log messages from the dashboard is very cumbersome. The UI and UX are acceptable for a 'My Passport Wireless'-type device, but, for something more complicated like the My Cloud Business Series NAS units, there is scope for improvement. The number of third-party apps available for the My Cloud OS is minimal right now. However, this will definitely improve as the days go by. From WD's side, the firmware needs more QA. In addition, some of the features that were touted at release (such as the daisy-chaining of units using the LAN port) find no mention in the user manual.

At $530 for a diskless unit, the pricing is definitely better than the Synology DS415+ or the QNAP TS-453 Pro. However, for the performance that is on offer, the $480 QNAP TS-451 and the $500 Seagate NAS Pro 4-bay also enter the list of units to be considered. Considered purely on performance or price, the WD My Cloud DL4100 is not an obvious choice. Similar to the Seagate NAS Pro units, the channel support as well as the bundling of hard drives / pre-configured nature of the units can make the My Cloud DL4100 worth considering.