WD Red and WD Se NAS HDD Enterprise RAID Report

We put the WD Red and the WD Se's head-to-head in RAID 0 and RAID 5 tests. Having trouble deciding which drive is best for your new NAS? Read on.

Published
Updated
9 minutes & 57 seconds read time

Introduction

WD Red and WD Se NAS HDD Enterprise RAID Report 01

As with any purchasing decision, there is always the question of whether or not to make the jump to the next level of performance. The advent of a new class of consumer HDDs geared for NAS usage has blurred the line between consumer and nearline/enterprise HDD's. We have conducted full product evaluations of both the WD Red and the WD Se as single units, and today we will test them in RAID configurations.

WD's NAS HDD market segmentation is a perfect example of the shrinking difference between top-of-the-line nearline/enterprise HDD's and consumer hardware. The WD Red series is geared for the SMB and SOHO crowd with 1-5 bay NAS units. For a small price premium, customers can acquire the higher rated WD SE series of HDDs, geared for SMB (Small/Medium Business) and large enterprise applications with 6-12 bays.

In many cases, users looking for the next level of performance from their 1-5 bay NAS units turn to the WD Se. Both the WD Se and the WD Red share many characteristics that begin with the SATA 6Gb/s connection. They also share the same 64MB of cache and a bit error rate of 1 per 10E14. Both HDD's are designed for 24/7 use, and the duty cycle of a NAS HDD experiences many transitions from various idle states to active use. This is reflected in the robust 300,000 cycles Load/Unload cycle ratings for both drives.

One of the immediate performance improvements when making the jump from the Red to the Se is the difference in platter speed. The Red utilizes WD's IntelliPower technology, which many speculate alters the speed between 5,400 and 5,900 RPM's. The 7,200 RPM advantage leads to enhanced speed from the Se, but also an increase in power consumption. The Red consumes a peak of 4.5 Watts active, while the Se has a higher active consumption of 9.5 Watts. The noise of NAS HDD's can become a bit of a distraction in home user environments, and the WD Red has a lower decibel rating of 24, in contrast to the 28 offered by the Se.

WD Red and WD Se NAS HDD Enterprise RAID Report 02

The Se has a higher tolerance to heavy workloads, with up to 180 TB a year falling within its endurance envelope. The Red supports 120-150 TB a year, which is fine for most home users. Interestingly enough, the Red actually has a higher MTBF rating of one million hours, while also having a much shorter warranty period of three years. The WD Se has an MTBF rating of 800,000 hours, yet a longer warranty period of five years.

The disparity between the MTBF and warranty periods from both drives likely stems from the test environments during qualification. In larger rack environments, which the Se is designed for, there is a tremendous increase in vibration and heat. Vibration is introduced by other drives in the rack, and with the intended purpose being 6-12 drive racks, the drives are subjected to more vibration and heat. The drives are also tested with their respective workloads, and with the Se having a higher 180TB rating, the drive endures a more stringent test regimen.

In similar environments, the Se will outlive the Red, as reflected in the warranty period. The warranty period of a NAS is important, many users simply want to buy several high-quality drives, configure them in the NAS, and forget them for the next several years. For many the higher warranty period of five years for the Se, and the enhanced speed, make them an attractive solution.

Today we will test the 4-drive RAID performance of the WD Se and the WD Red series head-to-head in both RAID 0 and RAID 5 configurations.

Test System and Methodology

WD Red and WD Se NAS HDD Enterprise RAID Report 03
WD Red and WD Se NAS HDD Enterprise RAID Report 04

We utilize a new approach to RAID storage testing with our Enterprise Test Bench, designed specifically to target long-term performance with a high level of granularity.

Many testing methods record peak and average measurements during the test period. These average values give a basic understanding of performance, but fall short in providing the clearest view possible of I/O QoS (Quality of Service).

'Average' results do little to indicate the performance variability experienced during actual deployment. The degree of variability is especially pertinent, as many applications can hang or lag as they wait for I/O requests to complete. This testing methodology illustrates performance variability, and includes average measurements, during the measurement window.

While executing a workload, all storage solutions deliver variable levels of performance. While this fluctuation is normal, the degree of variability is what separates enterprise storage solutions from typical client-side hardware. Providing ongoing measurements from our workloads with one-second reporting intervals illustrates product differentiation in relation to I/O QoS. Scatter charts give readers a basic understanding of I/O latency distribution without directly observing numerous graphs.

Consistent latency is the goal of every storage solution, and measurements such as Maximum Latency only illuminate the single longest I/O received during testing. This can be misleading, as a single 'outlying I/O' can skew the view of an otherwise superb solution. Standard Deviation measurements consider latency distribution, but do not always effectively illustrate I/O distribution with enough granularity to provide a clear picture of system performance. We also use latency plots to illustrate latency scaling under various workloads.

The first page of results will provide the 'key' to understanding and interpreting our new test methodology. In replicated environments, RAID 0 can be a compelling choice for bleeding edge performance. RAID 5 provides a layer of data security that protects from the loss of a drive. We are testing both RAID 0 and RAID 5 in this evaluation. This evaluation consists of tests with default settings conducted on a LSI 9271-8i with a direct connection. We conduct our tests over the full LBA range to allow each HDD to highlight its average performance. Short stroking can increase performance at the loss of capacity.

RAID 0 4K Random Read/Write

WD Red and WD Se NAS HDD Enterprise RAID Report 05

Each QD for every parameter tested includes 300 data points (five minutes of one second reports) to illustrate the degree of performance variability. The line for each QD represents the average speed reported during the five-minute interval.

4K random speed measurements are an important metric when comparing drive performance, as the hardest type of file access for any storage solution to master is small-file random. One of the most sought-after performance specifications, 4K random performance is a heavily marketed figure.

Both arrays provide outstanding performance, with the WD Se setting the tone early with a blistering 779 IOPS average at QD256. The WD Red trails with 582 IOPS at QD256.

WD Red and WD Se NAS HDD Enterprise RAID Report 06

The WD Se does not win by as large a margin with read activity, the average of 617 IOPS at QD256 beats the Red by only 50 IOPS.

WD Red and WD Se NAS HDD Enterprise RAID Report 07

Our write percentage testing illustrates the varying performance of each solution with mixed workloads. The 100% column to the right is a pure write workload of the 4K file size, and 0% represents a pure 4K read workload.

Both drives provide good performance with the mixed workloads, though they both hit a bump in the road with 90% write data suffering a dip in performance.

WD Red and WD Se NAS HDD Enterprise RAID Report 08

The Red loses in the latency test, with more of its access taking longer to complete.

RAID 0 128K Sequential Read/Write

WD Red and WD Se NAS HDD Enterprise RAID Report 09

The 128K sequential speeds reflect the maximum sequential throughput of the HDD using a file size encountered in normal NAS usage.

The WD Red array averages a tremendous 537 MB/s average, while the Se only beats it slightly with 610 MB/s in our sequential write testing.

WD Red and WD Se NAS HDD Enterprise RAID Report 10

The WD Se array averages 651 MB/s of sequential read speed, and the Red trails with 555 MB/s.

WD Red and WD Se NAS HDD Enterprise RAID Report 11

When we test mixed read/write performance the gap between the two RAID arrays closes significantly.

WD Red and WD Se NAS HDD Enterprise RAID Report 12

The WD Se array beats the array of WD Reds, with the main variation occurring in the 10-40ms range.

RAID 0 Database/OLTP and Fileserver

RAID 0 Database/OLTP

WD Red and WD Se NAS HDD Enterprise RAID Report 13

This test emulates Database and On-Line Transaction Processing (OLTP) workloads. OLTP is in essence the processing of transactions such as credit cards and high frequency trading in the financial sector. Databases are the bread and butter of many deployments. These are demanding 8K random workloads with a 66% read and 33% write distribution that can bring even the highest performing solutions down to earth.

The WD Se array averages 645 IOPS in this mixed workload test, while the Red array trails with 555 IOPS at QD256.

WD Red and WD Se NAS HDD Enterprise RAID Report 14

The WD Red actually beats out the Se array in the lower-end of the latency chart, with great results in the .6-.8ms and 1-2ms range.

RAID 0 Fileserver

WD Red and WD Se NAS HDD Enterprise RAID Report 15

The File Server profile represents typical file server workloads that are very relevant to NAS usage. This profile tests a wide variety of different file sizes simultaneously to simulate multiple users with an 80% read and 20% write distribution.

The WD Red again only trails slightly, with an impressive average of 689 IOPS at QD256, a result that is very close to the 791 IOPS from the WD Se array.

WD Red and WD Se NAS HDD Enterprise RAID Report 16

RAID 0 Emailserver and Webserver

RAID 0 Emailserver

WD Red and WD Se NAS HDD Enterprise RAID Report 17

The Emailserver profile is a very demanding 8K test with a 50% read and 50% write distribution. This application is indicative of the performance of the solution in heavy write workloads.

The Se array averages 684 IOPS at QD256, while the Red array averages 580 IOPS.

WD Red and WD Se NAS HDD Enterprise RAID Report 18

RAID 0 Webserver

The Red array again delivers excellent performance at the lower latency ranges.

WD Red and WD Se NAS HDD Enterprise RAID Report 19

The Webserver profile is a read-only test with a wide range of file sizes. Web servers are responsible for generating content for users to view over the internet, much like the very page you are reading. The speed of the underlying storage system has a massive impact on the speed and responsiveness of the NAS, and thus the end-user experience.

The WD Se array averages 575 IOPS to the 504 IOPS from the Red array.

WD Red and WD Se NAS HDD Enterprise RAID Report 20

RAID 5 4K Random Read/Write

WD Red and WD Se NAS HDD Enterprise RAID Report 21

Write performance takes a huge hit with the RAID 5 arrays, and while the Se experiences significant variability the array of Reds delivers a very tight performance profile. The Se clearly wins in this test with a much higher average of 582 IOPS compared to the 149 IOPS from the WD Reds.

WD Red and WD Se NAS HDD Enterprise RAID Report 22

The read performance is on par with the results obtained in RAID 0, with an average of 604 IOPS for the Se, and 543 IOPS at QD256 for the Red array.

WD Red and WD Se NAS HDD Enterprise RAID Report 23

Our write percentage testing illustrates the varying performance of each solution with mixed workloads. The 100% column to the right is a pure write workload of the 4K file size, and 0% represents a pure 4K read workload.

Both arrays experience a slope of degraded performance as we progress into heavier write workloads. This is an expected result of the RAID 5 write performance penalty.

WD Red and WD Se NAS HDD Enterprise RAID Report 24

The Se clearly has the upper hand in the latency scores for RAID 5 random writes.

RAID 5 128K Sequential Read/Write

WD Red and WD Se NAS HDD Enterprise RAID Report 25

The 128K sequential speeds reflect the maximum sequential throughput of the HDD using a file size encountered in normal NAS usage.

The WD Reds take a huge hit in their sequential write performance in RAID 5; this is truly telling of the major performance differences between these two arrays. The Reds average only 106 MB/s, while the Se chugs along at a blistering 483 MB/s.

WD Red and WD Se NAS HDD Enterprise RAID Report 26

The read performance is again largely unaffected by RAID 5, with the Se's averaging 470 MB/s, and the Red array averaging 416 MB/s.

WD Red and WD Se NAS HDD Enterprise RAID Report 27

The Se's easily outperforms the Reds as we mix in heavier write workloads.

WD Red and WD Se NAS HDD Enterprise RAID Report 28

The Se delivers all of its I/O in a nice tight latency range.

RAID 5 Database/OLTP and Fileserver

RAID 5 Database/OLTP

WD Red and WD Se NAS HDD Enterprise RAID Report 29

The Red's average 281 IOPS compared to the Se with their score of 397 IOPS.

WD Red and WD Se NAS HDD Enterprise RAID Report 30

RAID 5 Fileserver

WD Red and WD Se NAS HDD Enterprise RAID Report 31

The Red array averages 695 IOPS, while the Se averages an outstanding 790 IOPS.

WD Red and WD Se NAS HDD Enterprise RAID Report 32

RAID 5 Emailserver and Webserver

RAID 5 Emailserver

WD Red and WD Se NAS HDD Enterprise RAID Report 33

The Se array averages 349 IOPS, while the Red averages 265 IOPS at QD256.

WD Red and WD Se NAS HDD Enterprise RAID Report 34

RAID 5 Webserver

WD Red and WD Se NAS HDD Enterprise RAID Report 35

The Red array averages 504 IOPS, and the Se's average 349 IOPS at QD256.

WD Red and WD Se NAS HDD Enterprise RAID Report 36

Final Thoughts

WD Red and WD Se NAS HDD Enterprise RAID Report 37

Users looking for a more durable HDD for their NAS units desire as much reliability and performance as possible. Looking to a higher-tier HDD is a natural progression for those in the market for the best performance and reliability.

While the Red has plenty of durability, and is geared for demanding NAS workloads, the longer warranty period and resilience to heavier workloads will attract many to the WD Se. The higher spindle speed of the Se has the side effect of higher power consumption, but to users with more demanding workloads, this can be an acceptable trade-off for the increased performance.

You simply cannot get around the fact that most NAS units, especially in the 1-5 bay category, are constrained to the limit of a single gigabit Ethernet port. A single gigabit Ethernet port is limited to an effective speed of 120-125 MB/s. The sequential speeds we received from the RAID 0 array were in the 550-600 MB/s range from both arrays, which easily outstrips the speed of the Ethernet connection.

Where the real difference comes in is in multiple user environments, where all access becomes randomized. Users with heavy random access will find a large difference in performance with the faster Se series.

Most users will not be utilizing a RAID 0 in their NAS unless they dealing with replicated data protected by strong backup schemes. Even with the Red's pushing out 580 IOPS in our random tests, we have to keep in mind that is only a scant 2 MB/s. Moving up to the Se brought about gains to nearly 800 IOPS with a 4K random read workload. The gains were much higher in random read access than random writes, where we only observed an incremental increase in speed. With mixed read/write workloads, there isn't as much of a difference between the two solutions in RAID 0.

In RAID 5, the most common use case, the Se really comes into its own. With RAID 5 random write activity, the Se's blew past the Reds - it really was a no-contest. RAID 5 sequential performance is also penalized heavily for the Reds, while the Se array continued to chug along with great speed. Our tests with mixed read/write sequential workloads further illustrated the huge difference between the two solutions in RAID 5, with the Se coming out with a big win. When it came to our workload emulations, there was a much larger performance delta between the two arrays, with the Se coming away with a resounding win.

The performance was a big win for the Se where it matters, in the RAID 5 tests. The faster platter speeds and enterprise-class features shone through to win the day. Throwing an extra two years of warranty into the mix, for a total of five years, distances the Se from the Red that much more. With better reliability and performance, the Se makes a great NAS HDD for demanding environments.

PRICING: You can find products similar to this one for sale below.

USUnited States: Find other tech and computer products like this over at Amazon.com

UKUnited Kingdom: Find other tech and computer products like this over at Amazon.co.uk

AUAustralia: Find other tech and computer products like this over at Amazon.com.au

CACanada: Find other tech and computer products like this over at Amazon.ca

DEDeutschland: Finde andere Technik- und Computerprodukte wie dieses auf Amazon.de

The quest for benchmark world records led Paul further and further down the overclocking rabbit hole. SSDs and RAID controllers were a big part of that equation, allowing him to push performance to the bleeding edge. Finding the fastest and most extreme storage solutions led to experience with a myriad of high-end enterprise devices. Soon testing SSDs and Enterprise RAID controllers at the limits of their performance became Paul's real passion, one that is carried out through writing articles and reviews.

Newsletter Subscription
We openly invite the companies who provide us with review samples / who are mentioned or discussed to express their opinion. If any company representative wishes to respond, we will publish the response here. Please contact us if you wish to respond.