Mailing-List: contact user-help@couchdb.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@couchdb.apache.org
Received-SPF: pass (athena.apache.org: domain of johnlicheng@gmail.com
 designates 209.85.213.180 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CAJzZBAS9R9QrRT3_Gy2cpw1nCh0Uz49Z191-KENSkaw3uaaL+g@mail.gmail.com>
References: 
 <CAJzZBASjfdUVgmY14zCq9MC0+cf24yUzvfGhRAxoMOtw7TchMg@mail.gmail.com>
 <9E9A84DE-CA6F-453B-9FDC-C5BCA48F0B24@apache.org>
 <CAJzZBAS9R9QrRT3_Gy2cpw1nCh0Uz49Z191-KENSkaw3uaaL+g@mail.gmail.com>
From: John Cheng <johnlicheng@gmail.com>
Date: Thu, 11 Aug 2011 12:29:40 -0700
Message-ID: 
 <CAJzZBAQS_EjCMr7foFvzihsCYiuipNkHVcW43KkTtxq8dbacaQ@mail.gmail.com>
Subject: Re: CouchDB performance
To: user@couchdb.apache.org
Content-Type: multipart/alternative; boundary=001636ed6eb09dff3604aa3fd2d6

--001636ed6eb09dff3604aa3fd2d6
Content-Type: text/plain; charset=ISO-8859-1

On Thu, Aug 11, 2011 at 11:36 AM, John Cheng <johnlicheng@gmail.com> wrote:

> On Thu, Aug 11, 2011 at 11:02 AM, Adam Kocoloski <kocolosk@apache.org>wrote:
>
>> Hi John, first thing I'd point out is that Filipe and Paul have done quite
>> a lot of work on the JSON processor since 1.0.2 was released.  Trunk
>> contains a JSON processor that calls out to YAJL instead of relying on pure
>> Erlang code.  The improvement for large document bodies is pretty
>> substantial.  Also, processing MBs of JSON is going to be CPU-bound in
>> 1.0.2.  CouchDB makes good use of multicore servers to distribute some of
>> the load if you throw concurrent requests at it.  Hope that helps,
>>
>> Adam
>>
>
> It does help here. I'll see if I can get compile the latest version of
> CouchDB on Ubuntu 10 LTS. It's been an adventure trying to compile CouchDB
> so far!
>


Just compiled CouchDB 1.1.0

Testing with doc target size = "100 kB", # of properties = "10", updates =
"500", numDocs = "100"
each document is approximately 100.1 kB
Using increment handler function to perform update
500 updates in 42.1960351467 seconds, 11.8494545343 ops/second

Now the updates dropped from 23 updates/sec to 11.84 updates/second

Even inserts dropped to 24 inserts/second. Accroding to iostat, CPU was
mostly idle (~85 to 92%) and I'm only getting 300 kB writes per second.

I guess this could be a problem with httplib2. Is there any configuration or
compile time options I may have missed that could cause such a big drop in
performance?

-- 
---
John L Cheng

--001636ed6eb09dff3604aa3fd2d6--