| 1 |
On Mon, Oct 24, 2011 at 4:31 PM, Grant Edwards |
| 2 |
<grant.b.edwards@×××××.com> wrote: |
| 3 |
> On 2011-10-24, Florian Philipp <lists@×××××××××××.net> wrote: |
| 4 |
>> Am 24.10.2011 22:02, schrieb Grant Edwards: |
| 5 |
>>> On 2011-10-24, walt <w41ter@×××××.com> wrote: |
| 6 |
>>> |
| 7 |
>>>> I just bought an add-on USB3 adapter and outboard USB3/sata docking |
| 8 |
>>>> station, and I've been comparing the performance with my old e-sata |
| 9 |
>>>> outboard docking station. Not so good :( |
| 10 |
>>>> |
| 11 |
>>>> After getting some unreliable results with hdparm, I settled on |
| 12 |
>>>> copying one 3GB file from one partition of the outboard drive to |
| 13 |
>>>> another partition of the same drive. These results are highly |
| 14 |
>>>> reproducible, and favor e-sata over USB3 by a large margin. |
| 15 |
>>>> |
| 16 |
>>>> Over at least six trials on each docking station I consistently get |
| 17 |
>>>> 105 seconds for USB and 84 seconds for e-sata, a 5:4 ratio in favor |
| 18 |
>>>> of e-sata. |
| 19 |
>>> |
| 20 |
>>> Not surprising. Did you expect that adding a gateway device to the |
| 21 |
>>> communication path and another protocol layer on top of SATA would |
| 22 |
>>> make things faster? |
| 23 |
>>> |
| 24 |
>>>> I used the same hard disk and the same pci-e slot in the same |
| 25 |
>>>> minimally-loaded machine for all the runs, and got very consistent |
| 26 |
>>>> results every time. |
| 27 |
>>>> |
| 28 |
>>>> Basically, the USB3/sata docking station gets the same throughput as |
| 29 |
>>>> the older sata 1 drives connected to the onboard pci sata controller, |
| 30 |
>>>> which is still pretty respectable for an outboard drive, I think. |
| 31 |
>>> |
| 32 |
>>> Yep, SATA performs the same as SATA. AFAIK, eSATA and SATA are |
| 33 |
>>> identical apart from the physical specs for the connector, a few minor |
| 34 |
>>> voltage level differences (to imporove noise tolerance), and hot-plug |
| 35 |
>>> support. |
| 36 |
>> |
| 37 |
>> Normal SATA also offers hotplug. Usually works, too. |
| 38 |
> |
| 39 |
> I read somewhere that not all controllers support hotplug on |
| 40 |
> "internal" connectors, but I can't personally attest to having found |
| 41 |
> one that didn't. |
| 42 |
> |
| 43 |
>>>> So, has anyone out there done similar tests on USB3 drives yet? |
| 44 |
>>> |
| 45 |
>>> There are disk drives that talk USB3 natively and aren't just using |
| 46 |
>>> USB<->SATA gateways? |
| 47 |
>> |
| 48 |
>> Well, there is USB Attached SCSI (CONFIG_USB_UAS in the kernel). It |
| 49 |
>> supports command queuing and works for USB-2.0 and 3.0 (but has |
| 50 |
>> additional software overhead for USB-2.0). I've not yet seen a |
| 51 |
>> compatible device, though. |
| 52 |
> |
| 53 |
> Interesting. Is USB3 peer to peer like SCSI and Firewire, or is it |
| 54 |
> the same master/slave poll/response scheme that has always crippled |
| 55 |
> USB? Doing SCSI via a poll/response transport protocol seems like it |
| 56 |
> would lose most of the advantages of SCSI. |
| 57 |
|
| 58 |
IIRC USB3 is interrupt-driven instead of constantly polling the device. |
Archives