SSD cache copy slower than hard drive?


Recommended Posts

I use TeraCopy to transfer my files from my laptop to my UnRAID shares. and I was very pleased to see that transfer speeds were consistently around 11 MB/s.  However, after I installed an SSD cache, I noticed that the transfer speeds went actually down to 8-9 MB/s, when I expected them to go up... what is the use of a cache drive then?  Anyone see this before?

 

Link to comment

I use TeraCopy to transfer my files from my laptop to my UnRAID shares. and I was very pleased to see that transfer speeds were consistently around 11 MB/s.  However, after I installed an SSD cache, I noticed that the transfer speeds went actually down to 8-9 MB/s, when I expected them to go up... what is the use of a cache drive then?  Anyone see this before?

Transfer speeds that low suggest there is some other bottleneck in your system.  Even without a cache drive most people get around 30-40 MBps if on a wired GB LAN when transferring files to the parity protected array.  The speeds you quote are more typical of using a 100Mbps Lan or a wifi connection.

Link to comment

I use TeraCopy to transfer my files from my laptop to my UnRAID shares. and I was very pleased to see that transfer speeds were consistently around 11 MB/s.  However, after I installed an SSD cache, I noticed that the transfer speeds went actually down to 8-9 MB/s, when I expected them to go up... what is the use of a cache drive then?  Anyone see this before?

Transfer speeds that low suggest there is some other bottleneck in your system.  Even without a cache drive most people get around 30-40 MBps if on a wired GB LAN when transferring files to the parity protected array.  The speeds you quote are more typical of using a 100Mbps Lan or a wifi connection.

Or really.....really.....really old drives ;-)

Link to comment

I hadn't given it a thought that 11 MB/s was slow... now I have another thing to think about! I'm using brand-new HGST 3TB drives with a brand-new Samsung SSD on a wired powerline adapter Ethernet network; I should be getting lots higher speeds! And yes I had cache enabled on the shares because I saw the red triangles to the left of each share I was using.  I actually started using TeraCopy because using Windows Explorer was giving me less than 6 MB/s.  The ironic thing is that I can stream huge 1080p files without a hiccup!  I'm going to have to run more experiments and report back, this doesn't seem right at all.  Thanks for your help!

Link to comment

I hadn't given it a thought that 11 MB/s was slow... now I have another thing to think about! I'm using brand-new HGST 3TB drives with a brand-new Samsung SSD on a wired powerline adapter Ethernet network; I should be getting lots higher speeds! And yes I had cache enabled on the shares because I saw the red triangles to the left of each share I was using.  I actually started using TeraCopy because using Windows Explorer was giving me less than 6 MB/s.  The ironic thing is that I can stream huge 1080p files without a hiccup!  I'm going to have to run more experiments and report back, this doesn't seem right at all.  Thanks for your help!

 

The powerline part is most likely your number one limiting factor.

 

As for 1080p streaming, not so ironic. It doesn't require that much bandwidth.

Link to comment

Got it,  powerline ethernet is still twice as fast as my wifi network on a good day; I think I may need a new router.  Thanks again!

 

You don't actually need to replace your router.  You can just buy a Gigabit switch => you plug this into the router; then plug your other devices (PC, UnRAID Server,etc.) into the switch.    The devices connected to the switch will then have Gb connectively ... and instead of 11MB/s writes you should see 35-40MB/s writes; and your reads will be closer to 100MB/s.

 

Also, if you're using one of the newer powerline adapters, you should also see a significant increase in the speeds you get through that as long as you connect it to the Gb switch.  The newer 500Mb/s powerline units don't actually hit 500Mb in my experience, but I have seen transfer speeds in the 300Mb/s range ... which is nearly 40MBs.

 

Bottom line;  It's pretty clear that your main problem is you're using a 100Mb network => upgrade it to Gb and you'll see a very nice improvement.

 

Link to comment

The funny thing is that I do have a Megabit switch (see photo) connected to the Zyxel poweriine adapter, and where both sending and receiving devices connect, along with a few more.  I have no idea what's going on; I'll keep trying different things. It's not the end of the world, but it is quite puzzling.  Thanks for your help!

switch.jpg.9f1cc2ea284de79796e95ccaecc69b85.jpg

Link to comment

This is what I get:

 

login as: root

root@tower's password:

Last login: Tue Apr 28 12:15:17 2015 from 192.168.1.26

Linux 3.19.4-unRAID.

root@Tower:~# ethtool eth0

Settings for eth0:

        Supported ports: [ MII ]

        Supported link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

                                1000baseT/Full

        Supported pause frame use: No

        Supports auto-negotiation: Yes

        Advertised link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

        Advertised pause frame use: No

        Advertised auto-negotiation: Yes

        Speed: 100Mb/s

        Duplex: Full

        Port: MII

        PHYAD: 1

        Transceiver: external

        Auto-negotiation: on

        Supports Wake-on: g

        Wake-on: d

        Link detected: yes

root@Tower:~#

 

Any ideas?

Link to comment

MB/s is a megabyte per seconds.  Mb/s is a megabit per second. 

 

If the nic is connecting at 100 mb/s and transfers are going at 8 MB/s on a powerline connection, I'd think that was pretty OK. 

 

As to why it slowed down when cache was added I couldn't say.  My only thought on that is possibly change the sata cable, make sure it's rated for Sata 3 and not 1 or 2.  Could be a bad cable.  Same thing with the sata port. Could be a bad drive. 

 

When I was converting my drives to xfs, and was in the process of moving files around.  I noticed that one drive seemed to be a lot slower than the other drives during this process.  I connected it to a different available sata port (thankfully I had a port available on a different card) and it worked normally after that. 

Link to comment

This is what I get:

 

login as: root

root@tower's password:

Last login: Tue Apr 28 12:15:17 2015 from 192.168.1.26

Linux 3.19.4-unRAID.

root@Tower:~# ethtool eth0

Settings for eth0:

        Supported ports: [ MII ]

        Supported link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

                                1000baseT/Full

        Supported pause frame use: No

        Supports auto-negotiation: Yes

        Advertised link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

        Advertised pause frame use: No

        Advertised auto-negotiation: Yes

        Speed: 100Mb/s

        Duplex: Full

        Port: MII

        PHYAD: 1

        Transceiver: external

        Auto-negotiation: on

        Supports Wake-on: g

        Wake-on: d

        Link detected: yes

root@Tower:~#

 

Any ideas?

NIC supports 1000 but only connected at 100.

 

Check cables and ports at both ends.

Link to comment

This is what I get:

 

login as: root

root@tower's password:

Last login: Tue Apr 28 12:15:17 2015 from 192.168.1.26

Linux 3.19.4-unRAID.

root@Tower:~# ethtool eth0

Settings for eth0:

        Supported ports: [ MII ]

        Supported link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

                                1000baseT/Full

        Supported pause frame use: No

        Supports auto-negotiation: Yes

        Advertised link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

        Advertised pause frame use: No

        Advertised auto-negotiation: Yes

        Speed: 100Mb/s

        Duplex: Full

        Port: MII

        PHYAD: 1

        Transceiver: external

        Auto-negotiation: on

        Supports Wake-on: g

        Wake-on: d

        Link detected: yes

root@Tower:~#

 

Any ideas?

NIC supports 1000 but only connected at 100.

 

Check cables and ports at both ends.

And make sure all cables are Cat 6 or Cat 5e and not Cat 5.

Link to comment

This is what I get:

 

login as: root

root@tower's password:

Last login: Tue Apr 28 12:15:17 2015 from 192.168.1.26

Linux 3.19.4-unRAID.

root@Tower:~# ethtool eth0

Settings for eth0:

        Supported ports: [ MII ]

        Supported link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

                                1000baseT/Full

        Supported pause frame use: No

        Supports auto-negotiation: Yes

        Advertised link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

        Advertised pause frame use: No

        Advertised auto-negotiation: Yes

        Speed: 100Mb/s

        Duplex: Full

        Port: MII

        PHYAD: 1

        Transceiver: external

        Auto-negotiation: on

        Supports Wake-on: g

        Wake-on: d

        Link detected: yes

root@Tower:~#

 

Any ideas?

That shows the unRAID server is running at 100Mbps although the hardware is capable of running faster.    This means that some component (cabling or port on router/powerline) is only capable of supporting 100Mbps. 

 

It is surprising how often cabling comes into play as GB speeds require 8 cores in the cable to be working while 100MB only require 4 so a cable may 'appear' to be working but still limiting your speed.

Link to comment

Wait, who knew cables could make that much of a difference?  The server is an old tower I used for video editing back in 2008 (though it was a screamer back then).  I had that thing sitting around for years, so the sata cables must be from an old standard.  And the network cables definitely are Cat 5.  I'm surprised that it makes that much of a difference... live and learn!  I was waiting for the wife to go on a trip to get a new server  :)

 

In the meantime, on to Amazon to buy new cables.... thanks all for your help!

Link to comment

Wait, who knew cables could make that much of a difference?  The server is an old tower I used for video editing back in 2008 (though it was a screamer back then).  I had that thing sitting around for years, so the sata cables must be from an old standard.  And the network cables definitely are Cat 5.  I'm surprised that it makes that much of a difference... live and learn!  I was waiting for the wife to go on a trip to get a new server  :)

 

In the meantime, on to Amazon to buy new cables.... thanks all for your help!

Where I used to work, we had a lot of Cat 5 cabling in the walls.  They'd upgrade the switch in an area to one that was gigabit capable, and the next day we'd get a lot of calls from people whose nic in their pc's had auto-negotiated with the new switch gigabit speed and their network performance was terrible.  Hard set both ends to 100 Full and they were good to go.

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.