Return-Path: X-Original-To: apmail-helix-user-archive@minotaur.apache.org Delivered-To: apmail-helix-user-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id CA36B1064E for ; Sat, 29 Mar 2014 22:37:44 +0000 (UTC) Received: (qmail 55972 invoked by uid 500); 29 Mar 2014 22:37:44 -0000 Delivered-To: apmail-helix-user-archive@helix.apache.org Received: (qmail 55880 invoked by uid 500); 29 Mar 2014 22:37:43 -0000 Mailing-List: contact user-help@helix.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@helix.apache.org Delivered-To: mailing list user@helix.apache.org Received: (qmail 55866 invoked by uid 99); 29 Mar 2014 22:37:43 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 29 Mar 2014 22:37:43 +0000 X-ASF-Spam-Status: No, hits=1.5 required=5.0 tests=HTML_MESSAGE,RCVD_IN_DNSWL_LOW,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: domain of erwin.karbasi@gmail.com designates 209.85.216.175 as permitted sender) Received: from [209.85.216.175] (HELO mail-qc0-f175.google.com) (209.85.216.175) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 29 Mar 2014 22:37:37 +0000 Received: by mail-qc0-f175.google.com with SMTP id e16so7418969qcx.20 for ; Sat, 29 Mar 2014 15:37:15 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date:message-id:subject :from:to:content-type; bh=uKJYHdIwsvNKkJzHNDzy3mclb+tUdyQJERJR8xR+wYk=; b=qlPOdgZL5UzOQtmeho2z3gi8nLLrFQQ6Q9hFYfql44FkqhEMYUk5eyigOPqENUu5cn fL3Kj1GsZJCVYBCl4d76QCYVBaJea4gI8kKaVVH54ubs6U0gj3PSmtM20hcVABTLnqGn wxN6lTKBIP0jEpFwqWDKic9eC3jrEGXDlQeECBVBDuv7MsD3kXEWlM0+0ysxUkZ5lrTh zf9Zy62u6DF2xbeEitdGyP8mGdfH9MUk+vx6S61CEcdUJpEPKgwdgRk/oEByY+FKs2pS QNVu42iTjrt4pwcydSN7/PpQBlRKLQj5HPHFtjKaH2D14WFwyee4PcT2eImshEUAdq/s RtFg== MIME-Version: 1.0 X-Received: by 10.140.34.168 with SMTP id l37mr3568803qgl.85.1396132635703; Sat, 29 Mar 2014 15:37:15 -0700 (PDT) Sender: erwin.karbasi@gmail.com Received: by 10.140.83.20 with HTTP; Sat, 29 Mar 2014 15:37:15 -0700 (PDT) Received: by 10.140.83.20 with HTTP; Sat, 29 Mar 2014 15:37:15 -0700 (PDT) In-Reply-To: References: Date: Sun, 30 Mar 2014 00:37:15 +0200 X-Google-Sender-Auth: aFkSMJCKG1Xl_G3sy8YIHmCsyCw Message-ID: Subject: RE: Comparison question From: Erwin Karbasi To: user@helix.apache.org Content-Type: multipart/alternative; boundary=001a11c1092ecf7d7e04f5c67555 X-Virus-Checked: Checked by ClamAV on apache.org --001a11c1092ecf7d7e04f5c67555 Content-Type: text/plain; charset=ISO-8859-1 Thanks a lot for fast turnaround response. I agree with you but I found out that quorum in ZK can treat leader election mechanism by using ServerService class. But I didn't figure out how to get callback in the client side when leader elected or polling the quorum or one of the members in the quorum who is the leader. On Mar 30, 2014 12:42 AM, "Kanak Biscuitwala" wrote: > > If you do leader election with ZooKeeper, you basically have to try to > create have to follow the instructions in this recipe: > http://zookeeper.apache.org/doc/trunk/recipes.html#sc_leaderElection -- > this involves creating an ephemeral+sequential node, and listening for > changes to the nodes that precede yours. > > If you do leader election in Helix, you configure your resource and > participants up front, and then you're told by callback when you're leader, > and you're told by callback when you've given up leadership. If you have > multiple different logical entities that must have leaders, you can model > this by just adding another partition to a resource. It's less error-prone > this way, but you do need to spend extra time bringing up Helix controllers. > ________________________________ > > Date: Sat, 29 Mar 2014 23:05:46 +0200 > > Subject: Fwd: Re: Comparison question > > From: erwin@optinity.com > > To: user@helix.apache.org > > > > > >>> Hello Masters, > >> > >>> We had some internal discussion regarding the comparison among > > Helix and Zookeeper in term of leader election mechanism. > >>> > >>> I'd appreciate if you could elaborate in several points about the > > benefit of Helix against the ZooKeeper in term of Leader Election > > mechanism. What can I do in Helix that is missing in ZooKeeprt in term > > of Leader Election. > >>> > >>> Thanks in advance, > > Erwin > >> > > > > ---------- Forwarded message ---------- > > From: "Erwin Karbasi" > > > Date: Mar 30, 2014 12:02 AM > > Subject: Re: Comparison question > > To: "kishore g" >, > > user-subscribe@helix.incubator.apache.org>> > > Cc: > > > > > >> Hello Masters, > > > >> We had some internal discussion regarding the comparison among Helix > > and Zookeeper in term of leader election mechanism. > >> > >> I'd appreciate if you could elaborate in several points about the > > benefit of Helix against the ZooKeeper in term of Leader Election > > mechanism. What can I do in Helix that is missing in ZooKeeprt in term > > of Leader Election. > >> > >> Thanks in advance, > >> Erwin > --001a11c1092ecf7d7e04f5c67555 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable

Thanks a lot for fast turnaround response.
I agree with you but I found out that quorum in ZK can treat leader electio= n mechanism by using ServerService class. But I didn't figure out how t= o get callback in the client side when leader elected or polling the quorum= or one of the members in the quorum who is the leader.

On Mar 30, 2014 12:42 AM, "Kanak Biscuitwal= a" <kanak.b@hotmail.com&= gt; wrote:

If you do leader election with ZooKeeper, you basically have to try to crea= te have to follow the instructions in this recipe: = http://zookeeper.apache.org/doc/trunk/recipes.html#sc_leaderElection --= this involves creating an ephemeral+sequential node, and listening for cha= nges to the nodes that precede yours.

If you do leader election in Helix, you configure your resource and partici= pants up front, and then you're told by callback when you're leader= , and you're told by callback when you've given up leadership. If y= ou have multiple different logical entities that must have leaders, you can= model this by just adding another partition to a resource. It's less e= rror-prone this way, but you do need to spend extra time bringing up Helix = controllers.
________________________________
> Date: Sat, 29 Mar 2014 23:05:46 +0200
> Subject: Fwd: Re: Comparison question
> From: erwin@optinity.com
> To: user@helix.apache.org=
>
>
>>> Hello Masters,
>>
>>> We had some internal discussion regarding the comparison among=
> Helix and Zookeeper in term of leader election mechanism.
>>>
>>> I'd appreciate if you could elaborate in several points ab= out the
> benefit of Helix against the ZooKeeper in term of Leader Election
> mechanism. What can I do in Helix that is missing in ZooKeeprt in term=
> of Leader Election.
>>>
>>> Thanks in advance,
> Erwin
>>
>
> ---------- Forwarded message ----------
> From: "Erwin Karbasi" <erwin@optinity.com<mailto:= erwin@optinity.com>>
> Date: Mar 30, 2014 12:02 AM
> Subject: Re: Comparison question
> To: "kishore g" <g= .kishore@gmail.com<mailto:g.k= ishore@gmail.com>>,
> <user-= subscribe@helix.incubator.apache.org<mailto:user-subscribe@helix.incubator.apache.= org>>
> Cc:
>
>
>> Hello Masters,
>
>> We had some internal discussion regarding the comparison among Hel= ix
> and Zookeeper in term of leader election mechanism.
>>
>> I'd appreciate if you could elaborate in several points about = the
> benefit of Helix against the ZooKeeper in term of Leader Election
> mechanism. What can I do in Helix that is missing in ZooKeeprt in term=
> of Leader Election.
>>
>> Thanks in advance,
>> Erwin
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0
--001a11c1092ecf7d7e04f5c67555--