In message <[email protected]>
          Vincent Sanders <[email protected]> wrote:

> OK perhaps I needed to be more explicit in what I wanted to achieve
> from this testing.

Thanks.

> This means the browser must be used to retrieve numerous web pages

But if the "Cache disabled" message comes up, is everything visited 
after that irrelevant to the reported stats? On the ARMini, cache on 
the SD card, I can't browse for more than a couple of minutes before 
the warning comes up.

(2376.700000) content/llcache.c llcache_finalise 3361: Backing store 
wrote 414276 bytes in 0 ms average 3213 bytes/second

(2376.710000) image/image_cache.c image_cache_fini 428: Cache 
total/hit/miss/fail (counts) 1867/1728/139/0 (100%/92%/7%/0%)
(2376.710000) image/image_cache.c image_cache_fini 436: Cache 
total/hit/miss/fail (size) 1825363688/0/1779190876/0 
(100%/46172812%/0%/0%)

> For example if your cache has only written a hundred bytes of data the
> time it takes to do that will hugely affect the write rate. This is
> the cause of several "false positives" on the "write rate too slow"
> previously because a write of a hundred bytes that took a millisecond
> was considered as "too slow" instead of a rounding error!

Is the write rate that important anyway? Isn't the speed it can be 
read back more important for a cache?

Bryan.

-- 
RISC OS User Group Of London  -  http://www.rougol.jellybaby.net/
RISC OS London Show           -  http://www.riscoslondonshow.co.uk/

Reply via email to