Mailing-List: contact common-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: common-dev@hadoop.apache.org
Received-SPF: neutral (nike.apache.org: local policy)
Date: Tue, 9 Aug 2011 22:47:27 -0700
From: Konstantin Boudnik <cos@apache.org>
To: common-dev@hadoop.apache.org
Subject: Re: Derby with Hadoop --Why?
Message-ID: <20110810054727.GA28720@linspire.com>
References: 
 <CAJNLWYyCa=6uhn9eVX3f04JO7JhSgC-UpDJ5N5gHgPrJUbbSKg@mail.gmail.com>
 <COL117-W35952599D19A3104C6AD558F200@phx.gbl>
 <CAJNLWYzguoj+NrZ3iCUhcysyCObhiUXzAVy+=d2PJksHzrpvxw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: 
 <CAJNLWYzguoj+NrZ3iCUhcysyCObhiUXzAVy+=d2PJksHzrpvxw@mail.gmail.com>
User-Agent: Mutt/1.5.18 (2008-05-17)

Well, for once Hive uses Derby by default as its metastore.
What make you think that Hadoop project is using derby? 

Also, this question seems to belong to common-dev@ (Cc'ed) raher then general@
(Bcc'ed)

Cos

On Wed, Aug 10, 2011 at 11:07AM, Saravana Kumar wrote:
> Thanks For the Explanation but needs some clarity as well
> 
> Do you mean to say all the Information required to run a map/reduce job is
> effectively stored in derby. It means hadoop(not Ecosystem) uses Derby?
> 
> On Tue, Aug 9, 2011 at 5:52 PM, Michael Segel <michael_segel@hotmail.com>wrote:
> 
> >
> > Derby?
> >
> > First a little history...
> > Derby started out long ago as Cloudscape. Cloudscape was bought by
> > Informix. Informix was bought by IBM. IBM didn't understand Cloudscape and
> > decided to open source the project under APL. Hence Derby was born.
> >
> > Derby is an excellent lightweight 100% java database. So when you have a
> > Java framework, using Derby makes a lot of sense. Derby is used to persist
> > some environment information and I believe its used in part of some of the
> > unit testing.
> >
> > Where Derby has been replaced by MySQL is when someone wanted a multi-user
> > database and they were more comfortable with MySQL than they were with
> > Derby. (Hint: Derby can be started as an embedded single user database, or
> > as a multi-user database by changing its invocation at startup. ;-)
> >
> > So I would guess the initial reason to go with Derby was that its released
> > under APL and there were no licensing issues. ;-)
> >
> >
> > > Date: Tue, 9 Aug 2011 15:17:35 +0530
> > > Subject: Derby with Hadoop --Why?
> > > From: saravana.hadoop@gmail.com
> > > To: general@hadoop.apache.org
> >  >
> > > Hi
> > >
> > > What is the significance of Derby in Hadoop Project.
> > > Why people are using Derby along with Hadoop
> > >
> > > Regards
> > > Saravana Kumar.J
> >
> >