Mailing-List: contact dev-help@spark.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@spark.apache.org
Received-SPF: pass (athena.apache.org: domain of mengxr@gmail.com designates
 74.125.82.172 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CA+B-+fzod5i6dsJWaf7oz6758DY6i+Q0ui0J2TBOYwW4zQs+cw@mail.gmail.com>
References: 
 <CA+B-+fw8EkK1ECg=GPACiJUCMRS6C4vVmQC9dwAfpxvcoPDZtA@mail.gmail.com>
	<CAJgQjQ9y_B1LATjt6c1-PmmytBzf2R1XyFVhMayVK2Hay5stMQ@mail.gmail.com>
	<CA+B-+fxE2rMOf_uURn7OtF_HXpjgZFSRZWD4QS=ZDyyvT-HTAQ@mail.gmail.com>
	<CAJgQjQ9UWjURVD2U0uTUUc2DHH5-_VXQ=uH8mFYZvjLgCLSr9w@mail.gmail.com>
	<CA+B-+fx64WwiSK_ymuvEXMnw2sd+F5fCL3-mQiJbexqyJgzbww@mail.gmail.com>
	<CA+B-+fxircx6nrzHnhoMjDRzGFnY+R7xvjwU49GjkfT39vVs8A@mail.gmail.com>
	<CA+B-+fyO10rcr5bXUaxQOodotmy7njwJujzukzj-ZU8zT-oAuA@mail.gmail.com>
	<CAJgQjQ-b+me5O_ZRqty_OniGFLPai0AFTu5mnZrhGYG9s-ekbg@mail.gmail.com>
	<CA+B-+fyx_aACX5igiY+ibmk6y1OOCY0QmoRO7qpo2SY-+FqYcw@mail.gmail.com>
	<CAJgQjQ-wvbaE+QJ2Mngzpy=g5i7=4ym59JVrAMtijwv4uTbfPw@mail.gmail.com>
	<CA+B-+fxYLfEWx_+jzdfNO3DFa5dc4VXfEYpffpOT0xCV=yV4Ug@mail.gmail.com>
	<CAJgQjQ8KQRswOkedse2SXByU6r-XCgVZK2ftda-JKnjSu_r+QA@mail.gmail.com>
	<CA+B-+fz1DMVZz=vOU3GteUVUBNuFmo-O2gtcVm1-T-_L3u9Tow@mail.gmail.com>
	<CA+B-+fzod5i6dsJWaf7oz6758DY6i+Q0ui0J2TBOYwW4zQs+cw@mail.gmail.com>
Date: Mon, 7 Jul 2014 09:23:56 -0700
Message-ID: 
 <CAJgQjQ__bt67q7NTKh=vr8tTvbSkw1VXt_kB6RtpRWtt+i_-zQ@mail.gmail.com>
Subject: Re: Constraint Solver for Spark
From: Xiangrui Meng <mengxr@gmail.com>
To: dev@spark.apache.org
Content-Type: text/plain; charset=UTF-8

Hey Deb,

If your goal is to solve the subproblems in ALS, exploring sparsity
doesn't give you much benefit because the data is small and dense.
Porting either ECOS's or PDCO's implementation but using dense
representation should be sufficient. Feel free to open a JIRA and we
can move our discussion there.

Best,
Xiangrui

On Fri, Jul 4, 2014 at 10:19 AM, Debasish Das <debasish.das83@gmail.com> wrote:
> I looked further and realized that ECOS used a mex file while PDCO is using
> pure Matlab code. So the out-of-box runtime comparison is not fair.
>
> I am trying to generate PDCO C port. Like ECOS, PDCO also makes use of
> sparse support from Tim Davis.
>
> Thanks.
> Deb