Return-Path: X-Original-To: apmail-flink-issues-archive@minotaur.apache.org Delivered-To: apmail-flink-issues-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id D8B6F10B6D for ; Fri, 13 Feb 2015 10:55:21 +0000 (UTC) Received: (qmail 10957 invoked by uid 500); 13 Feb 2015 10:55:12 -0000 Delivered-To: apmail-flink-issues-archive@flink.apache.org Received: (qmail 10921 invoked by uid 500); 13 Feb 2015 10:55:12 -0000 Mailing-List: contact issues-help@flink.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@flink.apache.org Delivered-To: mailing list issues@flink.apache.org Received: (qmail 10911 invoked by uid 99); 13 Feb 2015 10:55:12 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 13 Feb 2015 10:55:12 +0000 X-ASF-Spam-Status: No, hits=-2000.0 required=5.0 tests=ALL_TRUSTED,T_RP_MATCHES_RCVD X-Spam-Check-By: apache.org Received: from [140.211.11.3] (HELO mail.apache.org) (140.211.11.3) by apache.org (qpsmtpd/0.29) with SMTP; Fri, 13 Feb 2015 10:54:51 +0000 Received: (qmail 9220 invoked by uid 99); 13 Feb 2015 10:54:48 -0000 Received: from git1-us-west.apache.org (HELO git1-us-west.apache.org) (140.211.11.23) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 13 Feb 2015 10:54:48 +0000 Received: by git1-us-west.apache.org (ASF Mail Server at git1-us-west.apache.org, from userid 33) id 26F49E028F; Fri, 13 Feb 2015 10:54:48 +0000 (UTC) From: StephanEwen To: issues@flink.incubator.apache.org Reply-To: issues@flink.incubator.apache.org References: In-Reply-To: Subject: [GitHub] flink pull request: [FLINK-785] Chained AllReduce / AllGroupReduce... Content-Type: text/plain Message-Id: <20150213105448.26F49E028F@git1-us-west.apache.org> Date: Fri, 13 Feb 2015 10:54:48 +0000 (UTC) X-Virus-Checked: Checked by ClamAV on apache.org Github user StephanEwen commented on the pull request: https://github.com/apache/flink/pull/370#issuecomment-74236667 The chained reducer variant makes more sense. The question is, what is cheaper - serialized exchange through the in-mem-channel, or the copying. My gut feeling is that the copy is a bit better, so it may be worth it. --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastructure@apache.org or file a JIRA ticket with INFRA. ---