[rescue] Sun 711

[George Adkins]

> > write at 150MB/sec, as long as it can read 300MB/sec...
>
> It is illusory.  You can only "read" 300MB/sec in your example if half of
> it is already cached in RAM (from a previous read).
>
Understood, except that this isn't.  it's a 1 gig file, which has not ben 
pre-read anywhere else, plus there are no other tasks going on.  An Idle 
system with 1 interactive shell executing the commands on an otherwise still 
disk subsystem.  No bursty traffic, no concurrent access of the disks.  I 
already thought about this and did the test under conditions where cacheing 
would not be anything other than the ordinary read-ahead one would get during 
the read process.

In this case, it's not an illusion, it's flat-out read speed.

> If your read access is bursty, you may see higher appearant read speeds.
> For example, you read for 1sec and get 150MB, then do something else for
> 1sec, then read for 1sec and get 300MB, etc.  In this case you were really
> performing readahead for the 1sec when you were otherwise crunching
> numbers, but your true I/O never exceeded 150MB/s.  You will never
> sustain[1] a read speed higher than the data rate to the platters.
>
> -James [TANSTAAFL]
>
> [1] For sufficiently high values of sustain.
> _______________________________________________
> rescue list - http://www.sunhelp.org/mailman/listinfo/rescue