From issues-return-160155-archive-asf-public=cust-asf.ponee.io@flink.apache.org Sat Mar 24 23:15:10 2018 Return-Path: X-Original-To: archive-asf-public@cust-asf.ponee.io Delivered-To: archive-asf-public@cust-asf.ponee.io Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by mx-eu-01.ponee.io (Postfix) with SMTP id 20E0A18064F for ; Sat, 24 Mar 2018 23:15:09 +0100 (CET) Received: (qmail 85329 invoked by uid 500); 24 Mar 2018 22:15:04 -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 85320 invoked by uid 99); 24 Mar 2018 22:15:04 -0000 Received: from pnap-us-west-generic-nat.apache.org (HELO spamd4-us-west.apache.org) (209.188.14.142) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 24 Mar 2018 22:15:04 +0000 Received: from localhost (localhost [127.0.0.1]) by spamd4-us-west.apache.org (ASF Mail Server at spamd4-us-west.apache.org) with ESMTP id AAA05C0147 for ; Sat, 24 Mar 2018 22:15:03 +0000 (UTC) X-Virus-Scanned: Debian amavisd-new at spamd4-us-west.apache.org X-Spam-Flag: NO X-Spam-Score: -109.511 X-Spam-Level: X-Spam-Status: No, score=-109.511 tagged_above=-999 required=6.31 tests=[ENV_AND_HDR_SPF_MATCH=-0.5, KAM_ASCII_DIVIDERS=0.8, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_SPF_WL=-7.5, USER_IN_WHITELIST=-100] autolearn=disabled Received: from mx1-lw-eu.apache.org ([10.40.0.8]) by localhost (spamd4-us-west.apache.org [10.40.0.11]) (amavisd-new, port 10024) with ESMTP id q5F6KLCkFxlB for ; Sat, 24 Mar 2018 22:15:02 +0000 (UTC) Received: from mailrelay1-us-west.apache.org (mailrelay1-us-west.apache.org [209.188.14.139]) by mx1-lw-eu.apache.org (ASF Mail Server at mx1-lw-eu.apache.org) with ESMTP id B39865F23E for ; Sat, 24 Mar 2018 22:15:01 +0000 (UTC) Received: from jira-lw-us.apache.org (unknown [207.244.88.139]) by mailrelay1-us-west.apache.org (ASF Mail Server at mailrelay1-us-west.apache.org) with ESMTP id C30C1E0DE0 for ; Sat, 24 Mar 2018 22:15:00 +0000 (UTC) Received: from jira-lw-us.apache.org (localhost [127.0.0.1]) by jira-lw-us.apache.org (ASF Mail Server at jira-lw-us.apache.org) with ESMTP id 26FB7214F3 for ; Sat, 24 Mar 2018 22:15:00 +0000 (UTC) Date: Sat, 24 Mar 2018 22:15:00 +0000 (UTC) From: "Steven Zhen Wu (JIRA)" To: issues@flink.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (FLINK-9061) S3 checkpoint data not partitioned well -- causes errors and poor performance MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/FLINK-9061?page=3Dcom.atlassian= .jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D1641= 2814#comment-16412814 ]=20 Steven Zhen Wu commented on FLINK-9061: --------------------------------------- [~jgrier] Yes, we want to contribute this back. We can probably partner wit= h you=C2=A0to get the change upstream. > S3 checkpoint data not partitioned well -- causes errors and poor perform= ance > -------------------------------------------------------------------------= ---- > > Key: FLINK-9061 > URL: https://issues.apache.org/jira/browse/FLINK-9061 > Project: Flink > Issue Type: Bug > Components: FileSystem, State Backends, Checkpointing > Affects Versions: 1.4.2 > Reporter: Jamie Grier > Priority: Critical > > I think we need to modify the way we write checkpoints to S3 for high-sca= le jobs (those with many total tasks).=C2=A0 The issue is that we are writi= ng all the checkpoint data under a common key prefix.=C2=A0 This is the wor= st case scenario for S3 performance since the key is used as a partition ke= y. > =C2=A0 > In the worst case checkpoints fail with a 500 status code coming back fro= m S3 and an internal error type of=C2=A0TooBusyException. > =C2=A0 > One possible solution would be to add a hook in the Flink filesystem code= that allows me to "rewrite" paths.=C2=A0 For example say I have the checkp= oint directory set to: > =C2=A0 > s3://bucket/flink/checkpoints > =C2=A0 > I would hook that and rewrite that path to: > =C2=A0 > s3://bucket/[HASH]/flink/checkpoints, where HASH is the hash of the origi= nal path > =C2=A0 > This would distribute the checkpoint write load around the S3 cluster eve= nly. > =C2=A0 > For reference: https://aws.amazon.com/premiumsupport/knowledge-center/s3-= bucket-performance-improve/ > =C2=A0 > Any other people hit this issue?=C2=A0 Any other ideas for solutions?=C2= =A0 This is a pretty serious problem for people trying to checkpoint to S3. > =C2=A0 > -Jamie > =C2=A0 -- This message was sent by Atlassian JIRA (v7.6.3#76005)