🚀 go-pugleaf

In article <DD1Gu9.KwH@news.cern.ch>
Fer de Jong <fer@dxcern.cern.ch> wrote:
> >Most operating systems like UNIX will allocate all unused memory for cache.
> >Hence, sometimes the system will 4M+ of cache. Now, if you have 1M of cache
> >on your controller, you have ZERO chance of finding data in the controller
> >after missing the data in the OS.
> 
> I wonder if this ZERO chance is actually true, as far as I understand it,
> sone of the caching controllers have pretty elaborate caching schemes
> implemented on an onboard CPU, I suppose you assume that they only
> implement a 'most recently accessed' caching scheme wich would justify
> your ZERO change statement.

Yes, you are correct.  I was assuming LRU algorithm being used.  I am
interested to hear from the cache controller companies about the smart
algorithms.  After all, when the controller only has physical sector
numbers to play with, it is not so easy to out-smart the application
programs.  If the controller understands the file system, it would be
a totally different matter.  But, what file system, DOS, UFS, RFS?
-- 
Y.P. Cheng,   ycheng@advansys.com
"We will be successful, because we won't accept failure."