Out of curiosity, what does zookeeper use for inter-node communications?
How big can it scale?
On Wed, Nov 9, 2011 at 1:35 AM, Yun Mao wrote:
> Row-level db locking is an internal mechanism for database to
> implement ACID transactions. It's not supposed to be exposed to users
> and implementatio
On Thu, Nov 24, 2011 at 4:07 AM, Sandy Walsh wrote:
> Each provisioning operation is generally sequential (and perhaps even
> preferred to be for debugging). The only real parallel activity is, as you
> mentioned, creating many instances or mass migration/resize/backup/etc.
It also seems implic
On Thu, Nov 24, 2011 at 6:14 AM, Michael Pittaro
wrote:
> 'Request Manager' is growing on meI think it describes the
> externally visible function without all the confusion around
> orchestration, and doesn't imply any specific implementation details.
"Event Manager"?
Since not everything it
'Request Manager' is growing on meI think it describes the
externally visible function without all the confusion around
orchestration, and doesn't imply any specific implementation details.
mike
On Wed, Nov 23, 2011 at 9:24 AM, Gary Kotton wrote:
> Hi,
> How about:- YAO - "Yet another orches
Uh, with US Thanksgiving it seems we'll be delaying this weeks meeting until
next week again.
That said, Jesse, comstud and myself met this week to talk about some tactical
plans for the orchestration and scaling efforts. We're going to write this up
for feedback next week.
Stay tuned!
-S
__
Hi,
How about:- YAO - "Yet another orchestrator"
I also like YASM
Thanks
Gary
-Original Message-
From: nova-orchestration-bounces+garyk=radware@lists.launchpad.net
[mailto:nova-orchestration-bounces+garyk=radware@lists.launchpad.net
] On Behalf Of Day, Phil
Sent: Wednesday, Novembe
I know - sometimes I think all the good names are already taken ;-)
Maybe we should go with "YASM - yet another state manager", or "Request
Manager" ?
Phil
-Original Message-
From: Sandy Walsh [mailto:sandy.wa...@rackspace.com]
Sent: 23 November 2011 17:10
To: Day, Phil; nova-orchestra
heh, when I hear Transaction Management I think databases, but I see your point
about State Management too.
I'm comfortable with either.
-S
From: Day, Phil [philip@hp.com]
Sent: Tuesday, November 22, 2011 2:00 PM
To: Sandy Walsh; nova-orchestration@l
Each provisioning operation is generally sequential (and perhaps even preferred
to be for debugging). The only real parallel activity is, as you mentioned,
creating many instances or mass migration/resize/backup/etc.
-S
From: nova-orchestration-bounces+s
Hi,
I think that this is also applicable for the management of many instance
- for example, shutdown, reboot, snapshots etc.
VMware has a feature called VMotion. Does OpenStack have an equivalent?
If so this too should be added to the list above.
Thanks
Gary
-Original Message-
From: nova-
What's the use cases to run tasks in parallel? One that's quite
obvious is to create many instances. The task to create a single
instance seems quite sequential to me. Do we have other use cases? Is
there a more complicated one, such as requiring parallel tasks to
communicate? Thanks,
Yun
--
Mai
11 matches
Mail list logo
