[vbox-dev] is virtstor being implemented?

[Achim Hasenmüller]

On Oct 19, 2010, at 09:15 , Huihong Luo wrote:

> I was playing Microsoft Virtual Server 2005 SP1 this morning, and they have SCSI Shunt Driver, a kind of virtstor, and it causes quite a difference when using it, so I was wondering there might be something. Now I know it must be due to their poor implementation on regular disk emulation.
>  
> There is one minor thing about fixed size disk creation, it seems to be quite slow. 10G VHD fixed size takes about 30 mins. Is this normal? Microsoft's VHD driver on windows 7 seems to be much faster. I know this isn't very important, as it's not used at runtime.
> 

We share some history with the Microsoft Hyper-V/Virtual Server team and this was actually something we learned back in 2004. The team back then chose the Adaptec SCSI controller and only later found out that it was not such a good choice as this adapter contains a microprocessor on board for which the device driver uploads firmware which then has to be executed for command processing. That was rather slow to virtualize and is the reason that a special "SCSI shunt driver" was developed which bypasses all this and just forwards SCSI request packets to the host in a Windows specific format for direct processing. Well, that's pretty much what our SATA + SCSI/SAS controllers do.

About the fixed disk creation code, I guess this is a subtle difference in how we perform the seek. There is probably a way to make it allocate free space using the NTFS sparse file support. I remember that there was a reason we didn't rely on sparse file allocation. Feel free to look at the code and come up with a patch :-)

Achim

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.virtualbox.org/pipermail/vbox-dev/attachments/20101019/2d5f451f/attachment.html>