4TB Seagate Faceoff - DX vs DM


Recommended Posts

Seagate currently makes 2 different 4TB drives, DX and DM.  From what I have read online, DX drives run faster and hotter. Since I got my hands on a few of both, and wasn't sure which ones were better, it only makes sense to run some performance tests. Hope this may help someone else decide which drive to buy.  No one kind of test can tell the whole story, so I ran several. 

 

The Contenders

[u]Model         RPM    Cache   Platters  Firmware  Notes      [/u]
ST4000DX000   7200   128MB?  5x800GB     CC44    Older Model
ST4000DM000   5900    64MB   4x1000GB    CC43    Newer Model

 

I bought mine from Costco, currently on sale through Sunday for $160:

http://lime-technology.com/forum/index.php?topic=25675.0

 

Test System

MB - ASRock Z77 Extreme 4

CPU - Intel G2020

Memory - 4GB 2x2GB

PS - Corsair HX750

Case - NZXT H2

unRAID 5.0rc11 Pro, fresh install

 

The drives I purchased were external, so I removed them from the cases.  I used 3 of each type of drive.  During each test, nothing else was running on the system.  The NZXT H2 case hase 8 forward loading drive bays with 2 120mm fans in front of the drives and one 120mm fan in the rear.  The case also has a fan speed controller, it was set on high.  One type of drives were tested, pulled and replaced with the other so during the tests there were only 3 drives in the case, in the same bottom three slots.  Ambient room temperature varied from 64-70F.  No errors in syslog during any tests.

 

My Tests

1.  Preclare disks

2.  hdparm -tT

3.  diskspeed.sh (Thanks to WeeboTech)

4.  Parity sync

5.  Parity check

6.  Copy 32.5GB file to share without Parity

7.  Copy 32.5GB file to share with Parity

8.  Parity sync w/fan controller on low

 

Test Result #1 - Preclear Disks

[u]     HH:MM:SS   (Pre, Zero, Post MB/s)   Disk  [/u]
DX   37:06:00       125, 138, 55         38.5C
DM   38:42:30       114, 130, 54         32.5C

 

Notes:

Results are average of 2 drives.  1 DX and 1 DM drive were precleared in different systems so those results are not included.  Preclear was performed in a different machine within an Antec 302 case.  Drives were precleared two at a time.

 

Observations:

DX drive 4.34% Faster, DM drive 18.46% Cooler

 

DX

==  ST4000DX000-1CL160    Z1Z059MP
== Last Cycle's Pre Read Time  : 8:48:46 (126 MB/s)
== Last Cycle's Zeroing time   : 7:58:44 (139 MB/s)
== Last Cycle's Post Read Time : 20:02:03 (55 MB/s)
== Last Cycle's Total Time     : 36:50:35
== Total Elapsed Time 36:50:35
== Disk Start Temperature: 24C
== Current Disk Temperature: 38C,

==  ST4000DX000-1CL160    Z1Z05MF9
== Last Cycle's Pre Read Time  : 8:56:56 (124 MB/s)
== Last Cycle's Zeroing time   : 8:06:55 (136 MB/s)
== Last Cycle's Post Read Time : 20:17:59 (54 MB/s)
== Last Cycle's Total Time     : 37:22:50
== Total Elapsed Time 37:22:50
== Disk Start Temperature: 25C
== Current Disk Temperature: 39C,

 

DM

==  ST4000DM000-1CD168    Z3000XC0
== Last Cycle's Pre Read Time  : 9:30:35 (116 MB/s)
== Last Cycle's Zeroing time   : 8:21:26 (132 MB/s)
== Last Cycle's Post Read Time : 20:10:08 (55 MB/s)
== Last Cycle's Total Time     : 38:03:10
== Total Elapsed Time 38:03:10
== Disk Start Temperature: 23C
== Current Disk Temperature: 30C,

==  ST4000DM000-1CD168    Z3000XBJ
== Last Cycle's Pre Read Time  : 9:52:33 (112 MB/s)
== Last Cycle's Zeroing time   : 8:38:03 (128 MB/s)
== Last Cycle's Post Read Time : 20:50:44 (53 MB/s)
== Last Cycle's Total Time     : 39:22:22
== Total Elapsed Time 39:22:22
== Disk Start Temperature: 24C
== Current Disk Temperature: 35C,

 

 

Test Result #2 - hdparm -tT

[u]      MB/s[/u]
DX   176.71 
DM   140.34

 

Notes:

Read test with average of 3 tests per drive.

 

Observations:

DX drive is 25.9% Faster.

 

DX

Timing buffered disk reads: 528 MB in  3.01 seconds = 175.46 MB/sec
Timing buffered disk reads: 528 MB in  3.01 seconds = 175.46 MB/sec
Timing buffered disk reads: 528 MB in  3.01 seconds = 175.46 MB/sec
Timing buffered disk reads: 532 MB in  3.00 seconds = 177.29 MB/sec
Timing buffered disk reads: 536 MB in  3.01 seconds = 178.12 MB/sec
Timing buffered disk reads: 536 MB in  3.01 seconds = 178.13 MB/sec
Timing buffered disk reads: 532 MB in  3.01 seconds = 176.70 MB/sec
Timing buffered disk reads: 532 MB in  3.01 seconds = 176.71 MB/sec
Timing buffered disk reads: 532 MB in  3.01 seconds = 176.71 MB/sec

 

 

DM

Timing buffered disk reads: 422 MB in  3.01 seconds = 140.12 MB/sec
Timing buffered disk reads: 422 MB in  3.01 seconds = 140.18 MB/sec
Timing buffered disk reads: 422 MB in  3.01 seconds = 140.24 MB/sec
Timing buffered disk reads: 418 MB in  3.01 seconds = 139.03 MB/sec
Timing buffered disk reads: 426 MB in  3.01 seconds = 141.65 MB/sec
Timing buffered disk reads: 426 MB in  3.01 seconds = 141.45 MB/sec
Timing buffered disk reads: 422 MB in  3.01 seconds = 140.34 MB/sec
Timing buffered disk reads: 422 MB in  3.01 seconds = 140.15 MB/sec
Timing buffered disk reads: 422 MB in  3.02 seconds = 139.94 MB/sec

 

Test Result #3 - diskspeed.sh

[u]      MB/s[/u]
DX   ~106 
DM   ~118

 

Notes:

Write test using dd to get through cache to see sustained writes.

 

Observations:

DM drive is 11.3% faster.  Interesting, I didn't expect that at all.

 

DX

writing 10240000000 bytes to: /mnt/disk1/testfile
737268736 bytes (737 MB) copied, 5.13705 s, 144 MB/s
1253184512 bytes (1.3 GB) copied, 10.057 s, 125 MB/s
1777341440 bytes (1.8 GB) copied, 15.457 s, 115 MB/s
2283529216 bytes (2.3 GB) copied, 20.107 s, 114 MB/s
2791892992 bytes (2.8 GB) copied, 25.1315 s, 111 MB/s
3286574080 bytes (3.3 GB) copied, 30.1969 s, 109 MB/s
3818030080 bytes (3.8 GB) copied, 35.1811 s, 109 MB/s
4379247616 bytes (4.4 GB) copied, 40.2055 s, 109 MB/s
4965186560 bytes (5.0 GB) copied, 45.2346 s, 110 MB/s
5491135488 bytes (5.5 GB) copied, 50.2562 s, 109 MB/s
5989725184 bytes (6.0 GB) copied, 55.2767 s, 108 MB/s
6489576448 bytes (6.5 GB) copied, 60.4267 s, 107 MB/s
6993245184 bytes (7.0 GB) copied, 65.5766 s, 107 MB/s
7514140672 bytes (7.5 GB) copied, 70.3458 s, 107 MB/s
8039453696 bytes (8.0 GB) copied, 75.3777 s, 107 MB/s
8576930816 bytes (8.6 GB) copied, 80.5708 s, 106 MB/s
9074910208 bytes (9.1 GB) copied, 85.6765 s, 106 MB/s
9659454464 bytes (9.7 GB) copied, 90.6107 s, 107 MB/s
10192655360 bytes (10 GB) copied, 95.5964 s, 107 MB/s
10240000000 bytes (10 GB) copied, 96.4381 s, 106 MB/s

writing 10240000000 bytes to: /mnt/disk2/testfile
614011904 bytes (614 MB) copied, 5.02322 s, 122 MB/s
1183740928 bytes (1.2 GB) copied, 10.3969 s, 114 MB/s
1700180992 bytes (1.7 GB) copied, 15.1411 s, 112 MB/s
2184924160 bytes (2.2 GB) copied, 20.0908 s, 109 MB/s
2720957440 bytes (2.7 GB) copied, 25.12 s, 108 MB/s
3228410880 bytes (3.2 GB) copied, 30.2368 s, 107 MB/s
3706005504 bytes (3.7 GB) copied, 35.1624 s, 105 MB/s
4257543168 bytes (4.3 GB) copied, 40.251 s, 106 MB/s
4755410944 bytes (4.8 GB) copied, 45.1965 s, 105 MB/s
5339218944 bytes (5.3 GB) copied, 50.5909 s, 106 MB/s
5847692288 bytes (5.8 GB) copied, 55.4366 s, 105 MB/s
6366725120 bytes (6.4 GB) copied, 60.5966 s, 105 MB/s
6858445824 bytes (6.9 GB) copied, 65.4365 s, 105 MB/s
7328440320 bytes (7.3 GB) copied, 70.3162 s, 104 MB/s
7849657344 bytes (7.8 GB) copied, 75.3465 s, 104 MB/s
8361890816 bytes (8.4 GB) copied, 80.3764 s, 104 MB/s
8869721088 bytes (8.9 GB) copied, 85.3953 s, 104 MB/s
9399530496 bytes (9.4 GB) copied, 90.7864 s, 104 MB/s
9865861120 bytes (9.9 GB) copied, 95.4505 s, 103 MB/s
10240000000 bytes (10 GB) copied, 99.0803 s, 103 MB/s

writing 10240000000 bytes to: /mnt/disk3/testfile
703415296 bytes (703 MB) copied, 5.37205 s, 131 MB/s
1208062976 bytes (1.2 GB) copied, 10.132 s, 119 MB/s
1740076032 bytes (1.7 GB) copied, 15.452 s, 113 MB/s
2266833920 bytes (2.3 GB) copied, 20.4619 s, 111 MB/s
2764949504 bytes (2.8 GB) copied, 25.1054 s, 110 MB/s
3304358912 bytes (3.3 GB) copied, 30.1319 s, 110 MB/s
3843116032 bytes (3.8 GB) copied, 35.1514 s, 109 MB/s
4463653888 bytes (4.5 GB) copied, 40.3718 s, 111 MB/s
4955834368 bytes (5.0 GB) copied, 45.1918 s, 110 MB/s
5467341824 bytes (5.5 GB) copied, 50.4117 s, 108 MB/s
5989203968 bytes (6.0 GB) copied, 55.233 s, 108 MB/s
6491413504 bytes (6.5 GB) copied, 60.2618 s, 108 MB/s
7053833216 bytes (7.1 GB) copied, 65.2855 s, 108 MB/s
7554097152 bytes (7.6 GB) copied, 70.3128 s, 107 MB/s
8182678528 bytes (8.2 GB) copied, 75.331 s, 109 MB/s
8639271936 bytes (8.6 GB) copied, 80.3514 s, 108 MB/s
9140839424 bytes (9.1 GB) copied, 85.3807 s, 107 MB/s
9694949376 bytes (9.7 GB) copied, 90.4016 s, 107 MB/s
10232783872 bytes (10 GB) copied, 95.7814 s, 107 MB/s
10240000000 bytes (10 GB) copied, 95.8226 s, 107 MB/s

 

DM

writing 10240000000 bytes to: /mnt/disk1/testfile
677327872 bytes (677 MB) copied, 5.02594 s, 135 MB/s
1201383424 bytes (1.2 GB) copied, 10.0401 s, 120 MB/s
1796383744 bytes (1.8 GB) copied, 15.0659 s, 119 MB/s
2365969408 bytes (2.4 GB) copied, 20.1458 s, 117 MB/s
3029384192 bytes (3.0 GB) copied, 25.1148 s, 121 MB/s
3596010496 bytes (3.6 GB) copied, 30.142 s, 119 MB/s
4254061568 bytes (4.3 GB) copied, 35.1613 s, 121 MB/s
4813768704 bytes (4.8 GB) copied, 40.1912 s, 120 MB/s
5389747200 bytes (5.4 GB) copied, 45.2819 s, 119 MB/s
6013658112 bytes (6.0 GB) copied, 50.3856 s, 119 MB/s
6620865536 bytes (6.6 GB) copied, 55.3756 s, 120 MB/s
7179343872 bytes (7.2 GB) copied, 60.2748 s, 119 MB/s
7787981824 bytes (7.8 GB) copied, 65.3051 s, 119 MB/s
8376259584 bytes (8.4 GB) copied, 70.6117 s, 119 MB/s
8960038912 bytes (9.0 GB) copied, 75.3472 s, 119 MB/s
9590207488 bytes (9.6 GB) copied, 80.5054 s, 119 MB/s
10195432448 bytes (10 GB) copied, 85.5154 s, 119 MB/s
10240000000 bytes (10 GB) copied, 85.7847 s, 119 MB/s

writing 10240000000 bytes to: /mnt/disk2/testfile
663016448 bytes (663 MB) copied, 5.0189 s, 132 MB/s
1206801408 bytes (1.2 GB) copied, 10.0457 s, 120 MB/s
1826780160 bytes (1.8 GB) copied, 15.0649 s, 121 MB/s
2446820352 bytes (2.4 GB) copied, 20.4252 s, 120 MB/s
2985231360 bytes (3.0 GB) copied, 25.409 s, 117 MB/s
3534519296 bytes (3.5 GB) copied, 30.1275 s, 117 MB/s
4094258176 bytes (4.1 GB) copied, 35.1506 s, 116 MB/s
4654068736 bytes (4.7 GB) copied, 40.1709 s, 116 MB/s
5245486080 bytes (5.2 GB) copied, 45.4288 s, 115 MB/s
5854938112 bytes (5.9 GB) copied, 50.305 s, 116 MB/s
6416282624 bytes (6.4 GB) copied, 55.2654 s, 116 MB/s
7078892544 bytes (7.1 GB) copied, 60.5187 s, 117 MB/s
7605841920 bytes (7.6 GB) copied, 65.2756 s, 117 MB/s
8156664832 bytes (8.2 GB) copied, 70.335 s, 116 MB/s
8740172800 bytes (8.7 GB) copied, 75.4386 s, 116 MB/s
9366520832 bytes (9.4 GB) copied, 80.4386 s, 116 MB/s
9902314496 bytes (9.9 GB) copied, 85.3775 s, 116 MB/s
10240000000 bytes (10 GB) copied, 88.2823 s, 116 MB/s

writing 10240000000 bytes to: /mnt/disk3/testfile
615412736 bytes (615 MB) copied, 5.02409 s, 122 MB/s
1207647232 bytes (1.2 GB) copied, 10.0529 s, 120 MB/s
1834034176 bytes (1.8 GB) copied, 15.0795 s, 122 MB/s
2446029824 bytes (2.4 GB) copied, 20.1075 s, 122 MB/s
3013985280 bytes (3.0 GB) copied, 25.2239 s, 119 MB/s
3601724416 bytes (3.6 GB) copied, 30.2389 s, 119 MB/s
4168383488 bytes (4.2 GB) copied, 35.1691 s, 119 MB/s
4743455744 bytes (4.7 GB) copied, 40.1909 s, 118 MB/s
5365126144 bytes (5.4 GB) copied, 45.4138 s, 118 MB/s
5988591616 bytes (6.0 GB) copied, 50.2329 s, 119 MB/s
6589264896 bytes (6.6 GB) copied, 55.2499 s, 119 MB/s
7191553024 bytes (7.2 GB) copied, 60.2695 s, 119 MB/s
7780857856 bytes (7.8 GB) copied, 65.2901 s, 119 MB/s
8347628544 bytes (8.3 GB) copied, 70.3095 s, 119 MB/s
8936391680 bytes (8.9 GB) copied, 75.4936 s, 118 MB/s
9563902976 bytes (9.6 GB) copied, 80.3536 s, 119 MB/s
10171790336 bytes (10 GB) copied, 85.439 s, 119 MB/s
10240000000 bytes (10 GB) copied, 86.0234 s, 119 MB/s

 

Test Result #4 - Parity Sync

[u]     HH:MM:SS   MB/s   Parity  Disk1  Disk2[/u]
DX   08:02:00   138.3   44C    38C    38C
DM   08:38:00   128.7   33C    30C    31C

 

Notes:

MB/s is average and drive temps are when finished.

 

Observations:

DX drive is 7.5% faster.  Huge difference in Temp here, DX data drives 24.6% warmer and the parity drive a whopping 33.3% hotter! 

 

 

Test Result #5 - Parity Check

[u]     HH:MM:SS   MB/s   Parity  Disk1  Disk2[/u]
DX   07:59:00   139.2   44C    39C    39C
DM   08:40:00   128.2   34C    30C    31C

 

Notes:

MB/s is average and drive temps are when finished.

 

Observations:

Just about the same as Parity sync above.  DX drive is 8.6% faster.  DX data drives 27.9% warmer and the DX parity drive 29.4% hotter!

 

Test Result #6 - Copy 32.5GB file to share without Parity

[u]     HH:MM:SS   MB/s   Disk1[/u]
DX   00:05:01   110.5   41C
DM   00:05:02   110.3   34C

 

Notes:  All tests were run 3 times from a Win7 machine copying a bluray file to a share on disk1 in the array and results are averaged.  Temp is from the end of the third file copy.

 

Observations:

Speeds are essentially the same with the DX drive running 20.6% warmer.

 

Test Result #7 - Copy 32.5GB file to share with Parity

[u]     HH:MM:SS   MB/s   Parity  Disk1[/u]
DX   00:11:36   47.8   43C    39C
DM   00:13:34   40.9   43C    31C

 

Notes:  All tests were run 2 times from a Win7 machine copying a bluray file to a share on disk1 in the array and results are averaged.  Temp is from the end of the second file copy.

 

Observations:

DX drive is 16.8% faster.  DX data drive 25.8% warmer and DX parity drive 26.5% warmer.

 

Test Result #8 - Parity sync w/fan controller on low

[u]     Parity  Disk1   Disk2[/u]
DX     53C    46C    46C
DM     38C    35C    36C

 

Notes:  When I was done with the other tests, I decided to add this one.  I intend to run my fan controller on low and wanted to see the difference between the drives in this configuration since it's more real world for me.  I didn't let it finish, I just wanted to get to steady-state temps so watched until the temp didn't change for 5 minutes.  For the DM drives this was about 30 minutes and for the DX drives it was about 45 minutes.

 

Observations:

HOLY HEATER BATMAN!!! 53C is HOT!  DX parity drive is 39.5% Hotter!  DX data drives are 29.6% Hotter!

 

 

Summary

Disk Reads:  DX drives are faster

Disk Writes:  DM and DX drives are about the same.

Temperature:  DM drives are MUCH cooler, by as much as 39%!!!

Parity Sync/Check:  DX drives are faster by 7.5%, I assume reads must weigh heavier than writes in terms of duration.

Copying Files to the array:  No difference (assuming you use a cache disk)

 

Conclusion

Which drive is better?  It depends, you decide.  :)  I want to reduce the noise of my systems, so heat is my enemy.  The NZXT H2 case is a new build and I want to make it as quiet as possible.  For me it's clearly the DM drives that I want in my system.    Parity times are in the ballpark, writes to cache are the same and it runs up to 39% cooler?  No brainer!

 

Link to comment

awesome post... for a parity drive, ... I suppose I would want a disk with fast write capability?

 

+quite the awesome post, thanks for the work and details.

 

If you read the data, the DX is a bit faster as parity drive, and the DM is the faster drive writing.

 

I was indifferent to the DX vs DM, but I would now suggest hunting for DMs (even for parity) in the unRAID application. You're looking for low cost storage with a bit a sensitivity to in home usage (heat/noise), DM wins. That small performance increase of the DX is not going to seen in daily operation, but the heat and possible fan noise will.

Link to comment

awesome post... for a parity drive, ... I suppose I would want a disk with fast write capability?

 

+quite the awesome post, thanks for the work and details.

 

Thanks!  It was fun testing the drives, I really enjoyed it. 

 

Personally, I am going to use the DM drives for everything, including parity.  I am going to return the DX drives to Costco, they just run WAY too hot for me.  I completely agree with c3, you're not going to feel the performance increase from the DX in daily operation of an unRAID server, but the heat and fans required to cool them you will for sure.

 

I am very pleased so far with the performance of the DM drives.  They are faster than any of the drives in my old array, even 7200rpm drives that run warmer.  I think it's a great balance of performance to heat.  Now I am just crossing my fingers since I bought several of them from the same batch.  :)  Ya just never know with new drives.

 

I currently have 7 of the DM drives preclearing in my new system with temps holding steady from 32-37C.  Build info here.

Link to comment
  • 2 weeks later...

It was reported that the internal 3TB version of the drive in these has a 1 year warranty - haven't checked the ones I have (3TB internal) to confirm this however.  Hopefully the external version will have a longer warranty.

 

It is two years according to Newegg.

 

http://www.newegg.com/Product/Product.aspx?Item=N82E16822148844

 

And just to be sure, I checked one of my ST3000DM001 drives

 

http://support.seagate.com/customer/en-US/warranty_validation.jsp?form=0

 

that I bought last November during that Black Friday week sale.  They say my warranty is good through Dec 2014.

Link to comment
  • 2 months later...
  • 2 months later...
  • 2 weeks later...

Wonder how long it will be before a 4TB variant of the western digital red drives, and at that point I'm curious how they will compare w/ these drives.

 

According to their announcements, WD will ship a 4TB version by the end of the year, and a 5TB version very shortly thereafter.    The 4TB Red will be a 1TB/platter unit (just like the 3TBs) ... so it should perform on par with the DMs (which are also 1TB/platter -- the DX's are only 800GB/platter).

 

The interesting question is whether the 5TB WD Reds will be 5 platter drives, or 4 1.25TB/platter units.  Hopefully the latter, as that would give them a nice performance boost !!

 

Link to comment
  • 2 years later...

And in the ultimate face off... %20 annual failure rates on the DX drives aka the Barracuda. So DM wins.

 

https://www.backblaze.com/blog/hard-drive-failure-rates-q2-2016/

This post has already been reported 3 times. "Necroposting" is totally allowed here, the post is on topic, and the link within the post is the same as in this thread so it is not spam.

 

Possibly the poster is just a drive-by who only registered to make this one post. Maybe he even has some axe to grind. But I am going to let it stand unless some other mod wants to bilge it.

Link to comment

And in the ultimate face off... %20 annual failure rates on the DX drives aka the Barracuda. So DM wins.

 

https://www.backblaze.com/blog/hard-drive-failure-rates-q2-2016/

This post has already been reported 3 times. "Necroposting" is totally allowed here, the post is on topic, and the link within the post is the same as in this thread so it is not spam.

 

Possibly the poster is just a drive-by who only registered to make this one post. Maybe he even has some axe to grind. But I am going to let it stand unless some other mod wants to bilge it.

 

I'm also leaning on the side of caution since it is hard drive/unRAID Storage related.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.