Re: optimize SQuid

Anthony Geoffron writes:
> Here is the result of calamaris perl script:
> As you'll see , there is a ratio between hit/miss
> really big !!
> Waht do you think about this result ?
>
> I need your help please.

  Help with what exactly? I'm afraid I don't understand what you're
asking.

> Anthony.
...
> report squid 1 Proxy-Report (03.Aug 99 16:20:02 - 03.Aug 99 17:00:00)
> # Incoming TCP-requests by status
> status request % kByte % sec
> kB/sec
> --------------------------------- -------- ------ --------- ------ ----
> -------
> HIT 8039 36.97 18697 17.16 0
> 11.86
> TCP_IMS_HIT 4073 18.73 1369 1.26 0
> 14.37
> TCP_HIT 3126 14.38 15303 14.04 0
> 14.37
> TCP_MEM_HIT 453 2.08 939 0.86 0
> 134.73
> TCP_REFRESH_HIT 326 1.50 993 0.91 1
> 2.43
> TCP_NEGATIVE_HIT 61 0.28 92 0.08 0
> 151.25
> MISS 13111 60.29 89442 82.07 3
> 2.01
...

  Do you suspect something is wrong with this? It looks like it's in
the range of results you should expect from a properly functioning
Squid cache. It's a few percent higher (17% hits by byte volume) than
I got in my very small-scale live test (13% hits by byte volume).

  All in all, it looks like it's working fairly well unless I'm missing
something. I don't think you're likely to get too much higher than
this with Squid in terms of hit ratios, though I'm sure someone will
correct me on this if I'm wrong. In theory adding something like a
cache "pre-feed" (e.g. Skycache) will improve this further, but I
haven't yet seen anyone report actual numbers on that.
  -- Clifton

-- 
 Clifton Royston  --  LavaNet Systems Architect --  cliftonr@lava.net
        "An absolute monarch would be absolutely wise and good.  
           But no man is strong enough to have no interest.  
             Therefore the best king would be Pure Chance.  
              It is Pure Chance that rules the Universe; 
          therefore, and only therefore, life is good." - AC