[geeks] The new IPC/LX, from Dell?

[Joshua Boyd]

On Thu, Nov 19, 2009 at 08:40:26PM -0500, Lionel Peterson wrote:
> On Nov 19, 2009, at 6:27 PM, Joshua Boyd <jdboyd at jdboyd.net> wrote:
>
>> eSATA isn't faster than SAS, even though SAS can talk to a lot of
>> drives, so without someone breaking open the firewire protocol and  
>> SATA
>> protocol documents, we don't know for certain that eSATA has less
>> protocol overhead than Firewire, we only know that eSATA has a much
>> higher clock rate.  For all I know, firewire could be more efficient
>> than SATA or SAS.
>
> I'm not talking protocol, I'm talking about the fact that SATA drives  
> connected to an eSATA port without translation, FW requires translation 
> because there is no such thing as a native FW device (they are all IDE, 
> SATA, or other technologies with a bridge device to interface with FW).
>
> A FW bridge device can have a better SATA/IDE controller than is on a  
> mainboard, but ia FW bridge will not be faster than that same improved  
> controller on the system bus.

A bridge of any sort does imply an increase in latency.  Increased
latency doesn't not always imply diminished bandwidth though, and
compared to the latency already in a disk, the latency a good firewire
controller and bridge chip adds is most likely inconsequential.  Also, I
would hope that a good imaging tool would mostly be insensitive to
latency.