Brian

HDD
Seagate Enterprise Capacity 6TB 3.5 HDD Review Discussion

10 posts in this topic

The Seagate Enterprise Capacity 6TB 3.5 HDDv4 is a well performing high capacity drive, a drive that can add up to an additional 50% capacity over current drives with a relatively low price point and all the interface and security features the enterprise wants.

Seagate Enterprise Capacity 6TB 3.5 HDD Review

Share this post


Link to post
Share on other sites

I don't like how this review talks about how the Helium drive uses less power than the conventional Seagate drive, but it doesn't include any data indicating that. Some power usage measurements/graphs should be included if you're going to say that.

Share this post


Link to post
Share on other sites

That is a very good point and we're evaluating how to provide more meaningful data for enterprise reviews like this. For now, we're simply referencing the specs of the drives, which for power tend to be accurate.

Share this post


Link to post
Share on other sites

Hello,

we are using those disks for a Ceph cluster that we have create. The problem is that we are getting very low performance compared to for example our SATA drives that we have to our workstations.

Our hardware is a Dell R730xd server. NO RAID configuration is applied, just spare disks. The OS is Ubuntu 12.04 x64

The results we are getting are below.

These are the results on my workstation's SATA disk

============================================================================
vtzan@vtzan-Desktop:/backup$ dd if=/dev/zero of=test.dat oflag=direct bs=1G count=1
1+0 records in
1+0 records out
1073741824 bytes (1,1 GB) copied, 10,4504 s, 103 MB/s
vtzan@vtzan-Desktop:/backup$ dd if=/dev/zero of=test.dat oflag=direct bs=4M count=256
256+0 records in
256+0 records out
1073741824 bytes (1,1 GB) copied, 7,97246 s, 135 MB/s
vtzan@vtzan-Desktop:/backup$ dd if=/dev/zero of=test.dat oflag=direct bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1,1 GB) copied, 8,16091 s, 132 MB/s
vtzan@vtzan-Desktop:/backup$ dd if=/dev/zero of=test.dat oflag=direct bs=256k count=4096
4096+0 records in
4096+0 records out
1073741824 bytes (1,1 GB) copied, 8,05868 s, 133 MB/s
vtzan@vtzan-Desktop:/backup$ dd if=/dev/zero of=test.dat oflag=direct bs=64k count=16384
16384+0 records in
16384+0 records out
1073741824 bytes (1,1 GB) copied, 8,26585 s, 130 MB/s
vtzan@vtzan-Desktop:/backup$ dd if=/dev/zero of=test.dat oflag=direct bs=4k count=262144
262144+0 records in
262144+0 records out
1073741824 bytes (1,1 GB) copied, 21,4378 s, 50,1 MB/s

============================================================================

Here is the results from our DELL R730xd server :

========================================================================
root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=1G count=1
1+0 records in
1+0 records out
1073741824 bytes (1.1 GB) copied, 5.55031 s, 193 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=4M count=256
256+0 records in
256+0 records out
1073741824 bytes (1.1 GB) copied, 6.97119 s, 154 MB/s
root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 14.9184 s, 72.0 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=256k count=4096
4096+0 records in
4096+0 records out
1073741824 bytes (1.1 GB) copied, 40.0724 s, 26.8 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=64k count=16384
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 147.296 s, 7.3 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=4k count=262144
^C41481+0 records in
41481+0 records out
169906176 bytes (170 MB) copied, 362.847 s, 468 kB/s
========================================================================

Any suggestion?

Thank you

Edited by johnvam

Share this post


Link to post
Share on other sites

What OS is the workstation? That last workload looks like 4k random write... Most drives will get pretty low with that. What IO scheduler is that system setup to?

Share this post


Link to post
Share on other sites

Hello,

we are using those disks for a Ceph cluster that we have create. The problem is that we are getting very low performance compared to for example our SATA drives that we have to our workstations.

Our hardware is a Dell R730xd server. NO RAID configuration is applied, just spare disks. The OS is Ubuntu 12.04 x64

The results we are getting are below.

Here is the results from our DELL R730xd server :

========================================================================

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=1G count=1

1+0 records in

1+0 records out

1073741824 bytes (1.1 GB) copied, 5.55031 s, 193 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=4M count=256

256+0 records in

256+0 records out

1073741824 bytes (1.1 GB) copied, 6.97119 s, 154 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=1M count=1024

1024+0 records in

1024+0 records out

1073741824 bytes (1.1 GB) copied, 14.9184 s, 72.0 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=256k count=4096

4096+0 records in

4096+0 records out

1073741824 bytes (1.1 GB) copied, 40.0724 s, 26.8 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=64k count=16384

16384+0 records in

16384+0 records out

1073741824 bytes (1.1 GB) copied, 147.296 s, 7.3 MB/s

root@node-11:~/test-disk# dd if=/dev/zero of=test.dat oflag=direct bs=4k count=262144

^C41481+0 records in

41481+0 records out

169906176 bytes (170 MB) copied, 362.847 s, 468 kB/s

========================================================================

Any suggestion?

Thank you

Given your performance scales down almost linearly with block size this looks like a storage stack/driver/adapter optimization problem. The disk is obviously capable of reaching full speed under the right conditions, so I'd start looking through iostat -x to see where the bottlenecks are happening.

Also, is write caching on?

P.S. Please don't cross-post the same question in multiple threads.

That last workload looks like 4k random write...

Yet it's 100% sequential.

Share this post


Link to post
Share on other sites

Has anyone test the average access time for this Enterprise drive or the ST6000DX000 desktop drive version of this? I am getting an access time of about 17.5ms on HD Tune Pro with the desktop drive, versus about 12ms on a Caviar Black 2TB. Not good at all. But the File benchmark and benchmark (STR test) it does very well on.

Also,

the drive parks it's heads in under 30 seconds of idle time and it takes at least 1 full second for the drive to begin to fulfill requests when the heads are parked because it does some sort of initialization almost every time! I can disable the Advanced Power Management (APM) every time the drive is plugged in, but that is annoying.

Edited by danwat1234

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now