Sorry - the change is still in code review:
http://codereview.waveprotocol.org/500001

On 18 January 2011 02:18, JeanV <[email protected]> wrote:

> Is this change available in the source code? Can we have access to
> this functionality? I've updated the source code but this
> functionality does not seem to be present.
>
> On Jan 17, 5:15 am, Alex North <[email protected]> wrote:
> > My most recent change loads all waves from storage into memory when the
> > server starts up. This is clearly not a long term solution, but a short
> term
> > way to get the system to behave similarly to how it does now. I'm going
> to
> > leave it in this state for now. There are at least two things that need
> > fixing before we can allow some waves to exist on disk only, not in RAM.
> >
> > - The WaveServer and WaveMap classes need to pass requests to list all
> wave
> > ids directly through to storage, so that the store can check the
> filesystem
> > (or whatever) without necessarily loading waves entirely into RAM in
> order
> > to enumerate them
> > - The index wave needs rework so as not to require initialisation from
> every
> > wave at start-up
> >
> > This latter most likely requires secondary storage for the index, which
> is
> > something Tad and I have been investigating.
> >
> > I'm not planning to fix these right now, but wanted to leave this note
> for
> > whenever someone gets round to it.
>
> --
> You received this message because you are subscribed to the Google Groups
> "Wave Protocol" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to
> [email protected]<wave-protocol%[email protected]>
> .
> For more options, visit this group at
> http://groups.google.com/group/wave-protocol?hl=en.
>
>

Reply via email to