jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nicolas " <nto...@gmail.com>
Subject Re: Alternative JSR 170 implementation?
Date Tue, 11 Jul 2006 10:20:16 GMT
Hi,

I am actually paid by my company to write a content repository scalable on
100 machines with fault tolerance. We were thinking of extending Jackrabbit
(everything would be committed back to Jackrabbit after). I am interested in
extending your solutions if it suits our needs.

Where can I find more information on what you built?

We are just starting the project and still in the requirements phase. (This
project is for my master's thesis.).

Nico
my blog! http://www.deviant-abstraction.net !!

On 7/11/06, Thomas Heute <theute@jboss.org> wrote:
>
>
> I would not recommend to start yet another project, you can simply change
> the
> persistence manager.
> The way we do it, is by using a singleton service with a distributed cache
> using a Hibernate persistent manager.
>
> If you are willing to improve our solution and serious about it feel free
> to
> contact me. (We would need to decouple it from our portal implementation
> first but this would not be hard since they are not highly coupled)
>
> Thomas.
>
> On Monday 10 July 2006 18:07, Sten R. Sandvik wrote:
> > Hi all.
> >
> >
> >
> > I must first say that Jackrabbit and JSR 170 is sure fun to work with.
> > But my professional experience so far is that Jackrabbit is not really
> > suited for large scale applications. Two features that the enterprise
> > market needs is clustering and online-backup support. I am therefore
> > polling the crowd for the use of another open-source project that has
> > the following main "features":
> >
> >
> >
> > (1)     Level 1 and 2 compliant. Optional features to be added when
> > level 1 and 2 is fully compliant.
> >
> > (2)     SQL-only data store. To simplify the implementation, Hibernate
> > or JPOX could be used. This takes care of online-backup, trasaction
> > support and clustering of datastore.
> >
> > (3)     Transaction aware caching. Plugin support (ehcache, oscache and
> > tangosol).
> >
> > (4)     Easy to configure with spring, especially spring 2.0.
> >
> >
> >
> > Is such an implementation needed? Anyone want to begin to work on such a
> > project? Any other open-source implementations out there that supports
> > clustering?
> >
> >
> >
> > Regards,
> >
> > Sten Roger Sandvik
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message