Re: Header Serialization Discussion

[Roberto Peon <[e-mail]>]

I also prefer just plain straight LRU. We can have some behavior which
allows for an item to be moved to the front of the LRU before it dies if
we're worried about it expiring and it is of high benefit.
(a flag on it saying it is 'permanent' or something as an option).

-=R


On Tue, Apr 16, 2013 at 11:37 AM, James M Snell <[e-mail]> wrote:

> On Tue, Apr 16, 2013 at 10:51 AM, James M Snell <[e-mail]> wrote:
> >[snip]
> > Requests:
> >                size  time | ratio min   max   std
> >      http1  200,876  0.01 | 1.00  1.00  1.00  0.00
> >        dhe   71,352  0.08 | 0.36  0.00  0.71  0.15
> >     delta2   35,850  0.37 | 0.18  0.03  0.67  0.12
> > headerdiff   38,673  0.23 | 0.19  0.03  0.88  0.16
> >
> > Responses:
> >                size  time | ratio min   max   std
> >      http1  160,435  0.06 | 1.00  1.00  1.00  0.00
> >        dhe   37,132  0.02 | 0.23  0.00  0.66  0.11
> >     delta2   42,764  0.43 | 0.27  0.02  0.65  0.10
> > headerdiff   46,321  0.69 | 0.29  0.04  0.84  0.13
> >
>
>
> I will note that these numbers reflect the fact that some header
> fields are intentionally set as ephemeral so that they are never
> stored in state. If I run the test again without that flag set for any
> of the fields the numbers come out as...
>
> REQUEST:
>                size  time | ratio min   max   std
>      http1  200,876  0.00 | 1.00  1.00  1.00  0.00
>        dhe   40,957  0.06 | 0.20  0.00  0.71  0.14
>     delta2   35,850  0.34 | 0.18  0.03  0.67  0.12
> headerdiff   38,673  0.29 | 0.19  0.03  0.88  0.16
>
> RESPONSE:
>                size  time | ratio min   max   std
>      http1  160,435  0.02 | 1.00  1.00  1.00  0.00
>        dhe   35,350  0.02 | 0.22  0.00  0.67  0.12
>     delta2   42,764  0.66 | 0.27  0.02  0.65  0.10
> headerdiff   46,321  0.57 | 0.29  0.04  0.84  0.13
>
> I still have more testing to do, of course, but that's where things stand
> now.
>
> - James
>
>