[rescue] emulation fun

[Mark Benson]

On Mon, 2009-02-02 at 09:22 -0500, Lionel Peterson wrote:
> Execution of code doesn't equate to capable of handling the same  
> workload. Mainframes have I/O throughput that PCs can't touch.

There's always one ;)

I didn't even dare to give the impression that a PC could carry that
kind of I/O load. For gods sakes mine falls on it's ass if it needs to
access 2 hard drives at once (and that's on SATA, no ATA). As for user
input our server copes okay with 10-12 users but it's finding it hard
going, and it's (as previously posted) not exactly a HP desktop
calculator. One thing that always bugs me is why PCs, and also is the
reason I think it's a shame specialist platforms like SGI and SPARC are
dying off, is still, in 2009, they can't make an x86 PC that multi-tasks
effectively. The OS is a big part, and for sure Linux and BSD cope way
better, but the whole architecture *still* sucks.

> It would be interesting to identify/calculate the lag where a PC can  
> equal the throughput of a particular mainframe model/series. (for  
> example, a current model Dell 2950 8-core Xeon, SAS drives equals the  
> performance and throughput of an IBM 3090/200

Sure would. I wouldn't have the first clue where to look, mind you, I'm
just a PC monkey ;)

-- 
Mark Benson
http://markbenson.org/blog