hadoop-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Dunning <ted.dunn...@gmail.com>
Subject Re: Announcing KeptCollections, distributed Java Collections for ZooKeeper
Date Thu, 09 Dec 2010 23:22:15 GMT
It definitely can go that way.

But some things are pretty nicely viewed that way.  For instance, wouldn't
it be nice to have a "live processes" set in different programs?  Of course,
you have to want watch for race conditions and you can't really believe it
is the same from moment to moment.  But with those caveats, it would be
*really* nice.

On Thu, Dec 9, 2010 at 1:27 PM, Adam Rosien <adam@rosien.net> wrote:

> Making a distributed system look like a collection is a very dubious
> proposition in my opinion. They are fundamentally different
> abstractions and reminds me of
> http://en.wikipedia.org/wiki/Fallacies_of_Distributed_Computing.
>
> .. Adam
>
> On Wed, Dec 8, 2010 at 12:40 PM, Anthony Urso <anthony.urso@gmail.com>
> wrote:
> > I am pleased to announce the initial release of KeptCollections, a
> > library of drop-in replacements for standard Java Collections that use
> > Apache ZooKeeper as a backing store.
> >
> > KeptCollections are designed to make it easy for anyone to write
> > distributed applications without having to learn the intricacies of
> > ZooKeeper, or distributed programming in general.
> >
> > The collections use the well-known JDK APIs, yet any changes made to
> > any of these collections by one node are seen by all other nodes
> > within milliseconds, allowing for easy communication between processes in
> a
> > computing cluster.
> >
> > More information here:
> >
> > https://github.com/anthonyu/KeptCollections/wiki
> >
> > and all code is available from:
> >
> > https://github.com/anthonyu/KeptCollections
> >
> > Please try it out, and let me know any problems you experience via
> > github issues or this email address.
> >
> > Cheers,
> > Anthony
> >
>

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