Mailing-List: contact derby-dev-help@db.apache.org; run by ezmlm
Precedence: bulk
Reply-To: <derby-dev@db.apache.org>
Received-SPF: pass (herse.apache.org: local policy)
Date: Tue, 24 Apr 2007 11:01:07 +0200
From: Knut Anders Hatlen <Knut.Hatlen@Sun.COM>
Subject: Re: 0.05% of test fixtures contribute 53% of the total running time!!
In-reply-to: <462DBA3D.8060005@sun.com>
Sender: Knut.Hatlen@Sun.COM
To: derby-dev@db.apache.org
Message-id: <wjo8k5w287u4.fsf@sun.com>
Organization: Sun Microsystems
MIME-version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-transfer-encoding: 8BIT
References: <462D84E4.8090802@apache.org> <462DAB7C.7010006@sun.com>
 <wjo8vefm8b48.fsf@sun.com> <462DBA3D.8060005@sun.com>
User-Agent: Gnus/5.110006 (No Gnus v0.6) Emacs/22.0.98 (usg-unix-v)

�ystein Gr�vlen <Oystein.Grovlen@Sun.COM> writes:

> Knut Anders Hatlen wrote:
>
>> I once wrote my own test runner which printed the time in milliseconds
>> for each test case, see the attached TimeRunner.java.
>
> Thanks, this is useful.
>
>>
>> Note that testPositionAgressive doesn't always take that much time. I
>> think that test and the other testPosition tests use random offset and
>> length and therefore the variation is high.
>
> While this is true for several of the test cases in BlobClob4BlobTest,
> by looking at the code, I am not able to find anything random in the
> execution of testPositionAgressive.  I will try to run them and see
> what I get.

I think you are correct, there's nothing random in that fixture. It
seems like it started taking a lot more time at revision 530085. I had
run the tests at an earlier revision.

-- 
Knut Anders