[Bioclusters] Re: file server for cluster

Ivo Grosse bioclusters@bioinformatics.org
Tue, 23 Apr 2002 13:53:15 -0400


Joe Landman <joe.landman@mscsoftware.com> wrote on Tue, 23 Apr 2002:

> Notation:
> 
>   Mb/s = Mega bit per second. 
>   MB/s = Mega byte per second (note the capitalized "b")
> 
>   1 bit = 0.125 byte, so 1 Mb/s -> 0.125 MB/s
> 
>   I would guess that most of what you quoted are MB/s, not Mb/s.
> 
>   from
> http://www.dell.com/us/en/esg/topics/segtopic_storage_scsi_main.htm you
> see the 160 MB/s throughput.  This corresponds to 1280 Mb/s.
> 
>   The conversion is not completely accurate as there are protocol and
> error detection/correction overheads.  I typically use a back of the
> envelope of a factor of 10 conversion.

Yes, and the web page I quoted uses Mb/s consistently.  Do you think 
it's a typo?

> It is going to be bandwidth limited by the number of PCI busses, the
> arrangement of IO, etc.  The specs say 3 PCI slots, but if they are on a
> single PCI bus, you will be able to swamp the PCI with either 1 Ultra160
> (for PCI-33), or 1 Ultra160+1 GigE card, or 2 Ultra160s.

Could you explain that (to a dummy)?

66 MHz x 64 bit = 4 Gb/s.

Note the "small" b.  :-)

Of course 4 Gb/s is purely theoretical.  What is the overhead?  20%?  
50%?

Even if it were 50%, that would leave 2 Gb/s?  How can that be 
saturated with 2 Ultra160s?  What is the w/r flux of one Ultra160?

Thanks!

Ivo