Return-Path: X-Original-To: apmail-cassandra-user-archive@www.apache.org Delivered-To: apmail-cassandra-user-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 47E8910DDD for ; Sat, 16 Nov 2013 19:19:49 +0000 (UTC) Received: (qmail 29313 invoked by uid 500); 16 Nov 2013 19:19:46 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 29269 invoked by uid 500); 16 Nov 2013 19:19:46 -0000 Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@cassandra.apache.org Delivered-To: mailing list user@cassandra.apache.org Received: (qmail 29261 invoked by uid 99); 16 Nov 2013 19:19:46 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 16 Nov 2013 19:19:46 +0000 X-ASF-Spam-Status: No, hits=1.5 required=5.0 tests=HTML_MESSAGE,MIME_QP_LONG_LINE,RCVD_IN_DNSWL_LOW,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: domain of mohitanchlia@gmail.com designates 209.85.160.53 as permitted sender) Received: from [209.85.160.53] (HELO mail-pb0-f53.google.com) (209.85.160.53) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 16 Nov 2013 19:19:39 +0000 Received: by mail-pb0-f53.google.com with SMTP id ma3so4964388pbc.26 for ; Sat, 16 Nov 2013 11:19:18 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=references:mime-version:in-reply-to:content-type :content-transfer-encoding:message-id:cc:from:subject:date:to; bh=8eM+6Nm+zBwYECuDIHTJK0r3RnFbBX3j7elJ29bekzU=; b=lnOYTbh+aUxYJ47IrwG2vo2AK4Yd6nApLVWpL4eaI4UEqNrsm2q4OzeLhaWSTwK3Ni CLwlUQutC1RFaYTa6VWsA6i15yXA/L5Iws+qyJSEq8ivVaBTCZKVgeul9jaRcuJ+cHn9 Ra0mC1wr9PuCJav2pBN6Nohjgk7oMT1H5jAddJXavpWbSJ6WqyuS93QYoqhnaz8L59EL VpEVEyMqjxfVnyzbJ1UdC71t9O1Xf+TbDyY39816Ay6tmjQChU34xvCFyceo3PTJam0E /218/BJIGKacqEFMSFLTf2LmMYP2Tuf789pOo9Pz3gqMVtLUPXLI13gi9J4HHKyL5iod o0iA== X-Received: by 10.68.133.198 with SMTP id pe6mr12889247pbb.10.1384629558094; Sat, 16 Nov 2013 11:19:18 -0800 (PST) Received: from [10.45.251.43] (mobile-198-228-223-044.mycingular.net. [198.228.223.44]) by mx.google.com with ESMTPSA id ja5sm12248038pbc.14.2013.11.16.11.19.17 for (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Sat, 16 Nov 2013 11:19:17 -0800 (PST) References: Mime-Version: 1.0 (1.0) In-Reply-To: Content-Type: multipart/alternative; boundary=Apple-Mail-B92C20AC-4237-473C-968C-A106E671FD80 Content-Transfer-Encoding: 7bit Message-Id: Cc: "user@cassandra.apache.org" X-Mailer: iPhone Mail (10B350) From: Mohit Anchlia Subject: Re: Commit log on USB flash disk? Date: Sat, 16 Nov 2013 11:14:13 -0800 To: "user@cassandra.apache.org" X-Virus-Checked: Checked by ClamAV on apache.org --Apple-Mail-B92C20AC-4237-473C-968C-A106E671FD80 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable In our testing USB tends to be slower. If there is something more integrated= internally would give you better performance Sent from my iPhone On Nov 16, 2013, at 8:30 AM, Dan Simpson wrote: > It doesn't seem like a great idea. The USB drives typically use dynamic w= ear leveling. See this analysis on wear: https://www.google.com/url?sa=3Dt&= rct=3Dj&q=3D&esrc=3Ds&source=3Dweb&cd=3D3&ved=3D0CD8QFjAC&url=3Dhttp%3A%2F%2= Fwww.usenix.org%2Fevent%2Ffast10%2Ftech%2Ffull_papers%2Fboboila.pdf&ei=3DqZy= HUrizFtKAygGs9YGoCg&usg=3DAFQjCNHTC7d6fcI1CNWmjbHMwgXI1nUWcQ&sig2=3DBaWgHj3i= b-cQOBPQsoCadA&bvm=3Dbv.56643336,d.aWc&cad=3Drjt >=20 > If you you do end up using it, make sure to monitor write latency so you d= on't get hit by the bus. >=20 >=20 > On Sat, Nov 16, 2013 at 6:12 AM, David Tinker wro= te: >> Our hosting provider has a cost effective server with 2 x 4TB disks >> with a 16G (or 64G) USB thumb drive option. Would it make sense to put >> the Cassandra commit log on the USB thumb disk and use RAID0 to use >> both 4TB disks for data (and Ubuntu 12.04)? >>=20 >> Anyone know how long USB flash disks last when use for a write heavy >> workload like this? >>=20 >> Please tell me if this is a really bad idea. >>=20 >> Our alternative is to use one 4TB disk for commit log and one for >> data. Of course this will give us only half the space. >>=20 >> Thanks >> David >=20 --Apple-Mail-B92C20AC-4237-473C-968C-A106E671FD80 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable
In our testing USB tends to be slower.= If there is something more integrated internally would give you better perf= ormance

Sent from my iPhone

On Nov 16, 2013, at 8:30 AM= , Dan Simpson <dan.simpson@gmail= .com> wrote:

It doesn't seem like a great idea.  The USB drives typically use d= ynamic wear leveling.  See this analysis on wear: https://www.google.com/url?sa=3D= t&rct=3Dj&q=3D&esrc=3Ds&source=3Dweb&cd=3D3&ved=3D0C= D8QFjAC&url=3Dhttp%3A%2F%2Fwww.usenix.org%2Fevent%2Ffast10%2Ftech%2Ffull= _papers%2Fboboila.pdf&ei=3DqZyHUrizFtKAygGs9YGoCg&usg=3DAFQjCNHTC7d6= fcI1CNWmjbHMwgXI1nUWcQ&sig2=3DBaWgHj3ib-cQOBPQsoCadA&bvm=3Dbv.566433= 36,d.aWc&cad=3Drjt

If you you do end up using it, make sure to monitor write lat= ency so you don't get hit by the bus.
=

On Sat, Nov 16, 2013 at 6:12 AM, David Ti= nker <david.tinker@gmail.com> wrote:
Our hosting provider has a cost effective serv= er with 2 x 4TB disks
with a 16G (or 64G) USB thumb drive option. Would it make sense to put
the Cassandra commit log on the USB thumb disk and use RAID0 to use
both 4TB disks for data (and Ubuntu 12.04)?

Anyone know how long USB flash disks last when use for a write heavy
workload like this?

Please tell me if this is a really bad idea.

Our alternative is to use one 4TB disk for commit log and one for
data. Of course this will give us only half the space.

Thanks
David

= --Apple-Mail-B92C20AC-4237-473C-968C-A106E671FD80--