Open Stack

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 02/06/2015 12:15 PM, Nikola Đipanov wrote:

> I've left a comment on the spec - basically I don't think this is an
> approach we should take.

Understood. There have been so many back-and-forth changes, with each
making the solution more and more complex, that it is hard to understand
the benefit. Essentially, we were trying to get as close to the eventual
state where the Scheduler knows everything it needs to make its decisions.

> Since I was not at the midcycle, I am sorry the discussions happened so
> close to the FF freeze, and there was not enough time to get broader
> feedback from the community in time.

No worries; it's not like there isn't anything else for you guys to be
reviewing or anything. :)

IAC, I've pushed a new revision of the spec. It's much simpler, with
much less impact, but it doesn't model the "scheduler knows all" ideal
as much. I'd be interested in your feedback, as well as everyone else's.
I think that this simpler approach will also be much more doable in the
Kilo timeframe.

- -- 

- -- Ed Leafe
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)
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=kh4g
-----END PGP SIGNATURE-----