Poor performance of WD Black

I purchased the WD Black 2013 versio 4 TB.  I found this drive to be considerably slower than 3 TB drives from other vendors such as Toshiba.  For example, if I use HD Tune, the read speed at the start is only between 151 to 171 MB/sec for a single drive, while other drive can easily hit 200 MB/sec.

I’m using SATA 3 connectors and wonder why the performance is suboptimal.  I’m using PC under Win 7 x64 in  a 3-drive software RAID 0 environment.  The RAID0 performance is only 330 - 350 MB/sec.  Please help me figure out performance is so poor even if I paid such a heavy premium for these drives.

Thanks -


Try running a diagnostic using WD DLG to make sure there is nothing wrong with the drive.

- [Data Lifeguard Diagnostic for Windows](http://support.wd.com/product/download.asp?groupid=616&sid=3&lang=en)

What does the benchmark curve look like?

The three disks are all okay.

Here is the becnhmark.  The first one is for 3 WD Black 2013 4 TB in software RAID0 of Win 7 x64.

WD BLACK 2013 


The second picture is for 3 Toshiba DT01ACA300 3TB in software RAID0 of Win 7 x64.  The Toshiba drive is less than 100 bucks, but WD 4 TB is about $ 250.  This makes me suspect that the premium I was paying is 5 year warranty and marketing expenses.  The performance may be artificially low in order to keep the failure rate in control.  I remember some online forum discuss this with regard to the RE enterprise drives.  I also heard that those benchmark drive sent to online commentators are different from consumers got.  



The single drive performance is similar:

WD HDTune_Benchmark_ATA _____ WDC_WD4003FZEX-0.png


HDTune_Benchmark_ATA _____ TOSHIBA_DT01ACA3.png

Sorry, it took some time for your images to be approved, but ISTM that the 4TB drive may have 5 platters whereas the 3TB drive may have 3. The lower data density would account for the lower transfer rate.

My rule of thumb is …

(transfer rate A) / (transfer rate B) = sqrt [(data density A) / (data density B)]

So (WD transfer rate) / (Toshiba transfer rate) = sqrt [800GB per platter / 1TB per platter]

That is, the WD drive’s transfer rate would be about 90% that of the Toshiba’s.

BTW, HD Tune is seeing only the first 2.2TB (2TiB) of each drive.

FYI, if you would like to determine the number of heads for each drive, see the following article:

How to determine number of heads using HD Tune:

Thank you for your help -

The Storage Review says the speed can read 224 MB/sec for the Black drive 4TB.  Is there anyway for me to achieve that?  I’m trying to see if such low performance is due to my setup is wrong.

Please show us the review, and please show us your shortstroked HD Tune benchmark graph, as per my tutorial.

Here is WD’s Black Series specification sheet:

Your WD4003FZEX is an Advanced Format model with a maximum sustained data transfer rate of 171 MB/s and a rotational speed 0f 7200 RPM.

The WD4001FAEX is the non-AF version with a transfer rate of 154 MB/s.

Therefore your drive is performing as specified.

Thank you for your analysis.  I assume that even if I get a hardware RAID card, I won’t see further performance gain for RAID 0 given that I achieved the maximum throughput of this drive?

I assume this is the review you are referring to:


The first thing I notice is that the reviewer has cherry-picked the specifications. He mentions the Buffer To Host transfer rate (6 Gbps) which is essentially useless, but he doesn’t mention the maximum sustained data transfer rate. The latter spec would probably place the WD models last in the comparison tests. Moreover, you need to ask yourself whether the “Enterprise Synthetic Workload Analysis” reflects your own usage patterns, or whether it is limited to server applications. For example, how relevant are the 4K read/write tests?

As for the figure of 224 MB/sec, the only reference I can find is 224 IOPS. That’s not the same thing.


IOPS = Input/Output Operations Per Second

The difference in the maximum sustained data transfer rate for the AF and non-AF WD drives appears to be directly related to the increase in the number of data bits per track. In a non-AF drive each 512-byte sector in a group of 8 would have its own overhead bytes, whereas in an AF drive a 4KB sector would have less combined overhead. This means that the remaining bits could be used for data storage rather than overhead. The typical improvement in data density is of the order of 10%.

171 / 154 = 1.11, ie an increase of 11%

To achieve the observed 48% performance increase, the drive relies on intelligent caching (“Dynamic Cache Technology”) and probably intelligent seeking. NCQ allows the drive to queue several I/O commands, and the drive could then optimise its seeking by accessing the closest sectors, thus minimising latency.

For example, instead of seeking to LBA 0 then LBA 1000 and then back to LBA 500, the drive could seek to LBA 0 then LBA 500 and then to LBA 1000. The other thing that all drives do is to cache an entire track in memory, the idea being that, if a head is sitting on a track, then the drive may as well retrieve all sectors within the track, not just the requested ones, in anticipation of the next read.

One other thing I notice is that the reviewer states that included in the benchmarks was a “128K (Sequential)” test, yet the graph shows the results for 128K 100% Read/Write Throughput. Can a “throughput” test really be viewed as a “sequential” test? Aren’t they measuring different things? AISI, a true sequential test would be limited by the rate at which data could be retrieved from the platters, rather than from cache, so a “sequential” result of 186.7MB/s doesn’t seem possible.

1 Like

Wow, thank you fzabkar for your comprehensive analysis.  I’m glad that my WD black is performing as it should be.  But hope it could be faster given the price premium I paid for the 4 TB version.

Thanks -

The performance limits of your drive are determined by the RPM, seek time, and the number of bits per track. These are fixed by the design. The best that you can do is to ensure that your most frequently accessed data are located on the fastest tracks.

If you examine a typical, fully stroked HD Tune benchmark graph, you will see that the transfer rate at the 0% mark is about twice the rate at the 100% mark. This is because the outermost track circumference is about twice that of the innermost tracks, and therefore the outer tracks pack about twice as many bits as the inner ones. The way to optimise your drive would be to partition it and confine your most frequently accessed data to the first partition. Not only will this maximise the transfer rate, but it will also reduce head movement, resulting in lower latency.

One other potential issue may be RAID alignment. This is not the same thing as 4KB alignment in an Advanced Format drive. Normally the standard Windows defaults are OK, but if you are using a striped RAID with a large stripe size of 2MB, say, then the standard partition offset of 1MB will result in your volumes straddling two stripes rather than one.

Thank you so much for your detailed analysis.  

Is there a software to re-arrange my data (without partitioning) so that my most valuable data is on the outmost part of my disk?  

Right now I have 3x 4 TB WD Black in my software RAID 0 array.  I’m debating whether to add the 4th drive to increase sequential I/O performance.  Yes, I’ll be very diligent in backing up my data.

I heard there is diminishing returns for adding more drives to the RAID 0.  Is it worthwhile?  Thanks -

try ultimate defrag

Sorry, other members would be better able to answer your RAID questions. I don’t have much experience in that area.

Thank you for your guys help !

glad you solved your issue

Like Apu says in the Simpsons

Thank you, Come again !!! lol