Return-Path: Delivered-To: apmail-hadoop-core-user-archive@www.apache.org Received: (qmail 99672 invoked from network); 3 Nov 2008 19:26:59 -0000 Received: from hermes.apache.org (HELO mail.apache.org) (140.211.11.2) by minotaur.apache.org with SMTP; 3 Nov 2008 19:26:59 -0000 Received: (qmail 59343 invoked by uid 500); 3 Nov 2008 19:27:00 -0000 Delivered-To: apmail-hadoop-core-user-archive@hadoop.apache.org Received: (qmail 58811 invoked by uid 500); 3 Nov 2008 19:26:59 -0000 Mailing-List: contact core-user-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: core-user@hadoop.apache.org Delivered-To: mailing list core-user@hadoop.apache.org Received: (qmail 58800 invoked by uid 99); 3 Nov 2008 19:26:59 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 03 Nov 2008 11:26:59 -0800 X-ASF-Spam-Status: No, hits=-2.0 required=10.0 tests=HTML_MESSAGE,RCVD_IN_DNSWL_MED,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (athena.apache.org: domain of pwyckoff@facebook.com designates 204.15.23.140 as permitted sender) Received: from [204.15.23.140] (HELO mailout-sf2p.facebook.com) (204.15.23.140) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 03 Nov 2008 19:25:43 +0000 Received: from mail.thefacebook.com (sc-hub02.thefacebook.com [192.168.18.105]) by pp01.sf2p.tfbnw.net (8.14.1/8.14.1) with ESMTP id mA3JQJ8t028699 for ; Mon, 3 Nov 2008 11:26:19 -0800 Received: from SC-MBXC1.TheFacebook.com ([192.168.18.100]) by sc-hub02.TheFacebook.com ([192.168.18.105]) with mapi; Mon, 3 Nov 2008 11:26:18 -0800 From: Pete Wyckoff To: "core-user@hadoop.apache.org" Date: Mon, 3 Nov 2008 11:26:17 -0800 Subject: Re: Status FUSE-Support of HDFS Thread-Topic: Status FUSE-Support of HDFS Thread-Index: Ack9o4+durLkyrgGSRSdhoZYYkLuqgARnoE/ Message-ID: In-Reply-To: <490ED9C8.4070103@apache.org> Accept-Language: en, en-US Content-Language: en X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en, en-US Content-Type: multipart/alternative; boundary="_000_C534905733478pwyckofffacebookcom_" MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=fsecure engine=1.12.7400:2.4.4,1.2.40,4.0.166 definitions=2008-11-03_07:2008-10-10,2008-11-03,2008-11-03 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 ipscore=0 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx engine=5.0.0-0805090000 definitions=main-0807290170 X-Virus-Checked: Checked by ClamAV on apache.org --_000_C534905733478pwyckofffacebookcom_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable +1 but since hadoop deals well with such directories currently, fuse-dfs wi= ll basically lock up on such directories - this is because ls -color=3Dblah= causes a stat on every file in a directory. There is a JIRA open for this= but it is a pretty rare case although it has happened to me at facebook. -- pete >It's good for a portable application to keep the #of files/directory low by having two levels of directory for storing files -just use a hash operation to determine which dir to store a specific file in. On 11/3/08 4:00 AM, "Steve Loughran" wrote: Pete Wyckoff wrote: > It has come a long way since 0.18 and facebook keeps our (0.17) dfs mount= ed via fuse and uses that for some operations. > > There have recently been some problems with fuse-dfs when used in a multi= threaded environment, but those have been fixed in 0.18.2 and 0.19. (do not= use 0.18 or 0.18.1) > > The current (known) issues are: > 2. When directories have 10s of thousands of files, performance can be = very poor. I've known other filesystems to top out at 64k-1 files per directory, even if they don't slow down. It's good for a portable application to keep the #of files/directory low by having two levels of directory for storing files -just use a hash operation to determine which dir to store a specific file in. --_000_C534905733478pwyckofffacebookcom_--