From commits-return-13851-archive-asf-public=cust-asf.ponee.io@hudi.apache.org Sat Mar 21 01:07:03 2020 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 [207.244.88.153]) by mx-eu-01.ponee.io (Postfix) with SMTP id 2CD7B18066D for ; Sat, 21 Mar 2020 02:07:03 +0100 (CET) Received: (qmail 60904 invoked by uid 500); 21 Mar 2020 01:07:02 -0000 Mailing-List: contact commits-help@hudi.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@hudi.apache.org Delivered-To: mailing list commits@hudi.apache.org Received: (qmail 60811 invoked by uid 99); 21 Mar 2020 01:07:02 -0000 Received: from mailrelay1-us-west.apache.org (HELO mailrelay1-us-west.apache.org) (209.188.14.139) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 21 Mar 2020 01:07:02 +0000 Received: from jira-he-de.apache.org (static.172.67.40.188.clients.your-server.de [188.40.67.172]) by mailrelay1-us-west.apache.org (ASF Mail Server at mailrelay1-us-west.apache.org) with ESMTP id 8AB8EE2F0E for ; Sat, 21 Mar 2020 01:07:01 +0000 (UTC) Received: from jira-he-de.apache.org (localhost.localdomain [127.0.0.1]) by jira-he-de.apache.org (ASF Mail Server at jira-he-de.apache.org) with ESMTP id 606097802D4 for ; Sat, 21 Mar 2020 01:07:00 +0000 (UTC) Date: Sat, 21 Mar 2020 01:07:00 +0000 (UTC) From: "Feichi Feng (Jira)" To: commits@hudi.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (HUDI-724) Parallelize GetSmallFiles For Partitions 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/HUDI-724?page=3Dcom.atlassian.j= ira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D170636= 97#comment-17063697 ]=20 Feichi Feng commented on HUDI-724: ---------------------------------- Hi [~vbalaji], partition touched depends on our data, the screenshot I atta= ched for "nogapAfterImprovement", it looks there are >=3D 45 partitions tou= ched(was hard-coded parallelism at that time). We do have a lot of files in= a partition(randomly checked one, it has 6000+ files, probably caused by r= ecordsize not been correctly set).=C2=A0 > Parallelize GetSmallFiles For Partitions > ---------------------------------------- > > Key: HUDI-724 > URL: https://issues.apache.org/jira/browse/HUDI-724 > Project: Apache Hudi (incubating) > Issue Type: Improvement > Components: Performance, Writer Core > Reporter: Feichi Feng > Priority: Major > Labels: pull-request-available > Attachments: gap.png, nogapAfterImprovement.png > > Original Estimate: 48h > Time Spent: 0.5h > Remaining Estimate: 47.5h > > When writing data, a gap was observed between spark stages. By tracking d= own where the time was spent on the spark driver, it's get-small-files oper= ation for partitions. > When creating the UpsertPartitioner and trying to assign insert records, = it uses a normal for-loop for get the list of small files for all partition= s that the load is going to load data to, and the process is very slow when= there are a lot of partitions to go through. While the operation is runnin= g on spark driver process, all other worker nodes are sitting idle waiting = for tasks. > For all those partitions, they don't affect each other, so the get-small-= files operations can be parallelized. The change I made is to pass the Java= SparkContext to the UpsertPartitioner, and create RDD for the partitions an= d eventually send the get small files operations to multiple tasks. > =C2=A0 > screenshot attached for=C2=A0 > the gap without the improvement > the spark stage with the improvement (no gap) -- This message was sent by Atlassian Jira (v8.3.4#803005)