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 1D348FFDD for ; Mon, 13 May 2013 20:57:36 +0000 (UTC) Received: (qmail 59281 invoked by uid 500); 13 May 2013 20:50:15 -0000 Delivered-To: apmail-cassandra-user-archive@cassandra.apache.org Received: (qmail 59252 invoked by uid 500); 13 May 2013 20:50:15 -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 59243 invoked by uid 99); 13 May 2013 20:50:15 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 13 May 2013 20:50:15 +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 edlinuxguru@gmail.com designates 209.85.212.172 as permitted sender) Received: from [209.85.212.172] (HELO mail-wi0-f172.google.com) (209.85.212.172) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 13 May 2013 20:50:09 +0000 Received: by mail-wi0-f172.google.com with SMTP id ey16so1365934wid.17 for ; Mon, 13 May 2013 13:49:49 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:content-type; bh=sebsnCX3YEM0l7yfowqjdnVKJHJJmCizR0WRFRcjvvY=; b=z6SfnGZso9bPyBT/JuOyMBfv9YKz7CuTZGWiuVAA5rQd3Nviksw1nrDtLlyp0vM/IG iYKd6GEyQPWa8Vq2tH7oTM+/BkNlk2tYkgqprP1UBxZuCLMCW8n4EADC5WDwWKIj8uoK TjtRCh8PatqJR50DSRgGjWSe0HVf2UBhg6wxx7qRi7xiPNczZpKdPbi9bod/nowY38XY I14JA8AOxCS1NSL4lt4YfUu/KBqNJFOts3iAU0i45nQ/xFDfIz4NiEXZttPk6gRcFXrd t4gabYiFNz+CjzLt1XPgVgTuaYbPpJWrLQAuueUq2XMEXI2HdAsL0PBFFOK134g+og6M 7jmA== MIME-Version: 1.0 X-Received: by 10.180.189.41 with SMTP id gf9mr22175721wic.32.1368478189109; Mon, 13 May 2013 13:49:49 -0700 (PDT) Received: by 10.194.20.164 with HTTP; Mon, 13 May 2013 13:49:49 -0700 (PDT) In-Reply-To: <518F41AD.8060307@filez.com> References: <518EDF67.3070709@filez.com> <518F41AD.8060307@filez.com> Date: Mon, 13 May 2013 16:49:49 -0400 Message-ID: Subject: Re: Astyanax client in Production From: Edward Capriolo To: "user@cassandra.apache.org" Content-Type: multipart/alternative; boundary=001a11c348305847cb04dc9fa8d7 X-Virus-Checked: Checked by ClamAV on apache.org --001a11c348305847cb04dc9fa8d7 Content-Type: text/plain; charset=ISO-8859-1 Being token aware makes a big performance difference. We do that internally with out client and it means a lot for 95 percentile time. If Astynax is not vnode token aware and your using them you could see worse performance. A long time "beef" with the client libraries is that they are always chasing cassandra. Astyanax looked awesome and sexy new compared to hector when it was released. But from what Radim is saying, it sounds like Astyanax has the same problem hector does. This is life in open source. I'm currently working on https://issues.apache.org/jira/browse/CASSANDRA-5435 because ...I need it.... That is what everything boils down to eventually. When netflix is ready to switch to vnodes they will update their client because ...they need it... :) On Sun, May 12, 2013 at 3:15 AM, Radim Kolar wrote: > Dne 12.5.2013 2:28, Techy Teck napsal(a): > >> I am running Cassandra 1.2.2 in production. What kind of problems you >> talking about? Might be I get some root cause why I am seeing bad read >> performance with Astyanax client in production cluster. >> > no support for full cassandra 1.2 feature set > no/bad support for virtual nodes > timeouts too often > --001a11c348305847cb04dc9fa8d7 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Being token aware makes a big performance difference.= We do that internally with out client and it means a lot for 95 percentile= time. If Astynax is not vnode token aware and your using them you could se= e worse performance.

A long time "beef" with the client libr= aries is that they are always chasing cassandra. Astyanax looked awesome an= d sexy new compared to hector when it was released. But from what Radim is = saying, it sounds like Astyanax has the same problem hector does.=A0

This is life in open source. I'm currently wo= rking on h= ttps://issues.apache.org/jira/browse/CASSANDRA-5435 because ...I need i= t.... That is what everything boils down to eventually. When netflix is rea= dy to switch to vnodes they will update their client because ...they need i= t... :)


On Sun,= May 12, 2013 at 3:15 AM, Radim Kolar <hsn@filez.com> wrote:
=
Dne 12.5.2013 2:28, Techy Teck napsal(a):
I am running Cassandra 1.2.2 in production. What kind of problems you talki= ng about? Might be I get some root cause why I am seeing bad read performan= ce with Astyanax client in production cluster.
no support for full cassandra 1.2 feature set
no/bad support for virtual nodes
timeouts too often

--001a11c348305847cb04dc9fa8d7--