maas-devel team mailing list archive

[John Arbash Meinel <john@xxxxxxxxxxxxxxxxx>]

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

...

>> I'll poke one more time, but being able to say "rebuilding tags
>> from 1 Region Controller to 20 Cluster Controllers" is network
>> bandwidth bound, rather than CPU bound or DB load bound is a
>> pretty good place to be.
> 
> Is rebuilding all tags at once across all machines a realistic
> scenario?  I would have thought that a more likely one is the
> introduction of a new rack (or racks) containing a few nodes at a
> time.
> 
> Cheers.
> 

I meant to reply to this as well. If you create a new tag, or change
the definition of a tag, it has to rebuild the node <=> tag mappings,
and it needs to process everything to do so.

A lot of what I'm doing is just trying to get a feel for how the
system works at scale, and I happen to have a specific problem that
I'm evaluating.

I think the other end of testing, "how many nodes can a single cluster
controller manage" would definitely be interesting to test. Though IMO
you need some actual machines for that. I wonder if you could get away
with 10,000 Raspberry Pi level machines. But even that would cost
$250,000, not to mention the networking infrastructure.

So really it would come down to a group that has hardware actually
doing something, and partnering with them to test scaling at that level.

John
=:->

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (Cygwin)
Comment: Using GnuPG with Mozilla - http://www.enigmail.net/

iEYEARECAAYFAlCA5usACgkQJdeBCYSNAAPemgCfZ2tMC1ov+gZMylSU2BNpa/qn
m1sAoK+JpOpWuPN11Ck9o56QsAzVu4RU
=LdIr
-----END PGP SIGNATURE-----