Date: Tue, 7 Dec 1999 21:48:09 +0100 From: Eivind Eklund <eivind@FreeBSD.ORG> To: Jeroen Ruigrok/Asmodai <asmodai@wxs.nl> Cc: cluster@FreeBSD.ORG Subject: Re: Homepage, take two. Message-ID: <19991207214809.I14851@bitbox.follo.net> In-Reply-To: <19991205133421.G2126@daemon.ninth-circle.org>; from asmodai@wxs.nl on Sun, Dec 05, 1999 at 01:34:21PM %2B0100 References: <19991201223902.E89541@daemon.ninth-circle.org> <19991205125437.J76366@bitbox.follo.net> <19991205133421.G2126@daemon.ninth-circle.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Dec 05, 1999 at 01:34:21PM +0100, Jeroen Ruigrok/Asmodai wrote:
> -On [19991205 13:00], Eivind Eklund (eivind@FreeBSD.ORG) wrote:
> >On Wed, Dec 01, 1999 at 10:39:02PM +0100, Jeroen Ruigrok/Asmodai wrote:
> >Rephrase of the first listed goal above, with extensions:
> >1. Turn BSD into a 'hot' clustering environment.
> >
> > This is our primary goal. In order to make this happen, we will
> > need to set a number of other goals, and reach for those. Below
> > are suggestions for subgoals, which again will need to be refined
> > to actually reach tasks to do.
> >
> >1. Create a focus for BSD clustering discussion
>
> cluster@freebsd.org list will be the focus of that discussion I gather?
This is one part of what we do for focusing. Another part is
documentation, and a place to start when trying to sort out BSD
clusters.
> >2. Make it easy to set up BSD based clusters.
2.1 Documentation
2.1.1 A guide to seeing what you need from a cluster, and choosing
available technology.
2.1.2 Guides for setting up each of the common cluster forms for BSD
2.2 Code
2.2.1 Pre-made ports for any relevant cluster packages
2.2.2 Writing any necessary tools that do not already exist
2.2.3 Integrate enabling technology into the BSD codebases
2.2.4 Integration of clustering in the BSD codebases, including a
'native' clustering system for BSD
> >3. Make it common to set up BSD based clusters.
This is a marketing goal, and depends on a combination of being close
to goal 2 and doing hype. Until we've made it reasonably easy (either
through docs or code) to set up clusters, I don't think there is any
point in pursuing this goal in any other way.
> >4. Make it natural to think of BSD when you want to set up a cluster.
This mixes closely with goal number 3, but also requires that we have
complete clustering solutions - ones that cover most of what people
need.
> >5. Make it natural to think of BSD when you want to do research on
> > clusters.
This meshes with goal 4, but further requires that it is easy to hook
into those parts of the system where it would be natural to research
on changes.
> >6. Make clusters a natural part of the 'toolchest' for BSD users.
This requires parts of point 2, and hype towards the present userbase.
> >7. Make clusters a part of the focus of the BSD developers.
This requires forward momentum, and some hype.
> >I'll get back to more refined tasks shortly.
See above.
> Please do. I stress that documentation should be a goal in itself. Or
> at least a subgoal of the above goals.
I see documentation as a means to an end (understanding/ability in the
reader), not as an ultimate goal in itself.
> >We need to have enough text that people
> >1) Understand what clusters are
> >2) Understand what high availability is (at least to a degree)
> >3) Understand why they would want clusters
> >4) Understand where there are 'leverage points' where they can
> > contribute to make a maximal difference
> >5) Understand why the route chosen forward (which we need to choose,
> > at least to a degree) has been chosen
>
> I think I can manage some initial text on 1 and 2, drawing from the
> texts already online.
I'll feed you with ideas for the other points, if you are willing to
write them up.
Eivind.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-cluster" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19991207214809.I14851>