Re: Rv: Why not BerkeleyDB based object store?

On Sat, Nov 29, 2008 at 1:01 AM, Lucas Brasilino
<lucas.brasilino_at_gmail.com> wrote:
> Hi Kinkie:
>
>> - workload is mostly writes
>> a well-tuned forward proxy will have a hit-rate of roughly 30%,
>> which means 3 writes for every read on average
>
> I don't know if I've misunderstood this statement.. Are you saying that
> even if an object is uncacheable, Squid stills storing it to disk ?

That was an oversimplification on my part, sorry about that.

-- 
    /kinkie