Return-Path: X-Original-To: apmail-hadoop-hdfs-issues-archive@minotaur.apache.org Delivered-To: apmail-hadoop-hdfs-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 5BDA0176CC for ; Mon, 13 Apr 2015 15:35:15 +0000 (UTC) Received: (qmail 25452 invoked by uid 500); 13 Apr 2015 15:35:15 -0000 Delivered-To: apmail-hadoop-hdfs-issues-archive@hadoop.apache.org Received: (qmail 25409 invoked by uid 500); 13 Apr 2015 15:35:15 -0000 Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: hdfs-issues@hadoop.apache.org Delivered-To: mailing list hdfs-issues@hadoop.apache.org Received: (qmail 25398 invoked by uid 99); 13 Apr 2015 15:35:15 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 13 Apr 2015 15:35:15 +0000 Date: Mon, 13 Apr 2015 15:35:14 +0000 (UTC) From: "Ming Ma (JIRA)" To: hdfs-issues@hadoop.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (HDFS-8009) Signal congestion on the DataNode MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/HDFS-8009?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14492523#comment-14492523 ] Ming Ma commented on HDFS-8009: ------------------------------- This overall DN congestion functionality seem useful. * System load is a useful metrics. But I wonder if we plan to take fairness into account. e.g., DN will ask only the heavy users to back off. * Support for read operation. Sometimes DNs could become the hotspot due to some applications reading specific blocks and cause high NIC bandwidth utilization. > Signal congestion on the DataNode > --------------------------------- > > Key: HDFS-8009 > URL: https://issues.apache.org/jira/browse/HDFS-8009 > Project: Hadoop HDFS > Issue Type: New Feature > Components: datanode > Reporter: Haohui Mai > Assignee: Haohui Mai > Fix For: 2.8.0 > > Attachments: HDFS-8009.000.patch > > > The DataNode should signal congestion (i.e. "I'm too busy") in the PipelineAck using the mechanism introduced in HDFS-7270. -- This message was sent by Atlassian JIRA (v6.3.4#6332)