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 CCF9F1023C for ; Mon, 2 Dec 2013 19:57:30 +0000 (UTC) Received: (qmail 67131 invoked by uid 500); 2 Dec 2013 19:57:28 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 67099 invoked by uid 500); 2 Dec 2013 19:57:28 -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 67091 invoked by uid 99); 2 Dec 2013 19:57:28 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 02 Dec 2013 19:57:28 +0000 X-ASF-Spam-Status: No, hits=1.5 required=5.0 tests=HTML_MESSAGE,RCVD_IN_DNSWL_LOW,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: domain of rcoli@eventbrite.com designates 209.85.220.172 as permitted sender) Received: from [209.85.220.172] (HELO mail-vc0-f172.google.com) (209.85.220.172) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 02 Dec 2013 19:57:22 +0000 Received: by mail-vc0-f172.google.com with SMTP id hz11so8925689vcb.3 for ; Mon, 02 Dec 2013 11:57:01 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:content-type; bh=3E0XHWzrmw+pw+bcR1iGCZbXaqI87ah3Uw/IXZq8JXs=; b=VNQEhIcgO9QfxOq4ltwjKTKKy/QZtgJXnM6uh5SYww+mPOLs76brA6rO/I3zlkfQXG SVpIbADNkhCp/l7VAtl4u9rmTB1WRI83o6ZYCW2K5YmnZKUMk9i28/K5ICI5v05vBw0u F9rzhi+UAz4EH0CN8T8R3TfRH2sJK8BpiuqubD8zRfr2qRd5OWG7oxxnhWQ0pGZcGsCv 0podh7xONDubie07WkLQIOTdBICrNvMoD/9w1xmbjLWfpDUw4gRMRKII5kqychQ+q2pq byWX+pR3XvF42ZR15VM5uZG5Q7AwWapduON/5tk2egKpTTai6f5tyHrmNbV73Vx9XsNV YU4w== X-Gm-Message-State: ALoCoQkPYda0dM13i3AR/5MJB3VMXPu6Q6vXWvaQhgFz+O1gEMB/SNHYMZBjBNR5zrPGzb95KCfc MIME-Version: 1.0 X-Received: by 10.58.22.33 with SMTP id a1mr758039vef.55.1386014221451; Mon, 02 Dec 2013 11:57:01 -0800 (PST) Received: by 10.220.150.3 with HTTP; Mon, 2 Dec 2013 11:57:01 -0800 (PST) In-Reply-To: References: Date: Mon, 2 Dec 2013 11:57:01 -0800 Message-ID: Subject: Re: Recommended amount of free disk space for compaction From: Robert Coli To: "user@cassandra.apache.org" Content-Type: multipart/alternative; boundary=089e011763a352d93a04ec9295b7 X-Virus-Checked: Checked by ClamAV on apache.org --089e011763a352d93a04ec9295b7 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable On Thu, Nov 28, 2013 at 7:21 PM, Robert Wille wrote: > So here=92s my question. If Cassandra only compacts one table at a time, > then I should be safe if I keep as much free space as there is data in th= e > largest table. If Cassandra can compact multiple tables simultaneously, > then it seems that I need as much free space as all the tables put > together, which means no more than 50% utilization. > If the number of SSTables is [x], then : - "minor" compaction runs on 1-[x-1] SSTables (in practice, usually a smaller number depending on config) - "major" compaction runs on [x] SSTables Unless you run a major compaction manually while using Size Tiered Compaction, you are very unlikely to run a major compaction in normal operation. Level compaciton strategy has a different set of assumptions. =3DRob --089e011763a352d93a04ec9295b7 Content-Type: text/html; charset=windows-1252 Content-Transfer-Encoding: quoted-printable
On Thu, Nov 28, 2013 at 7:21 PM, Robert Wille <rwille@fold= 3.com> wrote:
So here=92s my question. If Cass= andra only compacts one table at a time, then I should be safe if I keep as= much free space as there is data in the largest table. If Cassandra can co= mpact multiple tables simultaneously, then it seems that I need as much fre= e space as all the tables put together, which means no more than 50% utiliz= ation.

If the number of SSTables is [x], th= en :

- "minor" compaction runs on 1-[x-1= ] SSTables (in practice, usually a smaller number depending on config)
- "major" compaction runs on [x] SSTables

Unless you run a major compaction manually while using Size Tiered C= ompaction, you are very unlikely to run a major compaction in normal operat= ion.

Level compaciton strategy has a different set of assump= tions.

=3DRob =A0
--089e011763a352d93a04ec9295b7--