Return-Path: Delivered-To: apmail-cassandra-user-archive@www.apache.org Received: (qmail 57239 invoked from network); 7 Sep 2010 23:46:22 -0000 Received: from unknown (HELO mail.apache.org) (140.211.11.3) by 140.211.11.9 with SMTP; 7 Sep 2010 23:46:22 -0000 Received: (qmail 15380 invoked by uid 500); 7 Sep 2010 23:46:20 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 15340 invoked by uid 500); 7 Sep 2010 23:46:20 -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 15332 invoked by uid 99); 7 Sep 2010 23:46:20 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 07 Sep 2010 23:46:20 +0000 X-ASF-Spam-Status: No, hits=0.7 required=10.0 tests=SPF_NEUTRAL X-Spam-Check-By: apache.org Received-SPF: neutral (athena.apache.org: local policy) Received: from [209.85.215.172] (HELO mail-ey0-f172.google.com) (209.85.215.172) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 07 Sep 2010 23:46:16 +0000 Received: by eyd10 with SMTP id 10so3043537eyd.31 for ; Tue, 07 Sep 2010 16:45:54 -0700 (PDT) Received: by 10.216.188.20 with SMTP id z20mr6706wem.51.1283903148407; Tue, 07 Sep 2010 16:45:48 -0700 (PDT) MIME-Version: 1.0 Received: by 10.216.3.129 with HTTP; Tue, 7 Sep 2010 16:41:52 -0700 (PDT) In-Reply-To: References: <4C804E8E.1020105@softwareprojects.com> From: Benjamin Black Date: Tue, 7 Sep 2010 16:41:52 -0700 Message-ID: Subject: Re: 4k keyspaces... Maybe we're doing it wrong? To: user@cassandra.apache.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Really depends on your memtable settings and your memory. Larger memtables are generally better, so having lots of active ones works against making them as large as you might want. For example, if the flush MB threshold is 256 (which is the low end of what I recommend), even 50 of them could consume over 10GB. b On Tue, Sep 7, 2010 at 9:32 AM, Ned Wolpert wro= te: > This might be a loaded question... what is a 'lot of CFs'? 50? 250? 500? = =A0I > say loaded because I know that its probably dependent on how many keys > (size) are in a column family. > > On Mon, Sep 6, 2010 at 7:53 PM, Benjamin Black wrote: >> >> On Mon, Sep 6, 2010 at 12:41 AM, Janne Jalkanen >> wrote: >> > >> > So if I read this right, using lots of CF's is also a Bad Idea(tm)? >> > >> >> Yes, lots of CFs is bad means lots of CFs is also bad. > > > > -- > Virtually, Ned Wolpert > > "Settle thy studies, Faustus, and begin..."=A0=A0 --Marlowe >