Mailing-List: contact cassandra-user-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: cassandra-user@incubator.apache.org
Received-SPF: unknown (nike.apache.org: error in processing during lookup of
 timo.nentwig@wooga.net)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Apple Message framework v1077)
Subject: Re: Very simple benchmark - are this typical numbers?
From: Timo Nentwig <timo.nentwig@wooga.net>
In-Reply-To: <e06563880912071023m3c1b5c2exc1560952b1ad2661@mail.gmail.com>
Date: Tue, 8 Dec 2009 17:16:04 +0100
Content-Transfer-Encoding: quoted-printable
Message-Id: <93290460-A0BF-4501-8F52-9FA95B7A9549@wooga.net>
References: <CFE63079-75DE-4C87-AAB8-387C87780CDF@wooga.net>
  <e06563880912070859q5891bfa1k5ce199d7145821ab@mail.gmail.com>
  <58E59B99-799C-4A82-B446-E92E6C68CF33@wooga.net>
 <e06563880912071023m3c1b5c2exc1560952b1ad2661@mail.gmail.com>
To: cassandra-user@incubator.apache.org

On Dec 7, 2009, at 7:23 PM, Jonathan Ellis wrote:

> same thing, you are going to need multiple threads to max it out

I created up to 100 threads and read randomly. Some speed up but not =
actually mentionable. The threads didn't load the CPU mentionably =
either. I noticed that the thread dump was full of Thift =
(TBinaryProtocol something) InputStream.read(), 10MiB constant read from =
HD.

> but yes, reads are typically slower than writes in cassandra because
> of how the log-based merge structures work
>=20
> On Mon, Dec 7, 2009 at 11:09 AM, Timo Nentwig <timo.nentwig@wooga.net> =
wrote:
>>=20
>> On Dec 7, 2009, at 5:59 PM, Jonathan Ellis wrote:
>>=20
>>> yes and no -- that's about 4200/s, which is typical for only a =
single
>>=20
>> When writing, yes. But I would expect reading to be much faster (?). =
Re-executing the read test doesn't speed up things either (I/O caches).
>>=20
>>> thread but 1/3 to 1/5 of what you'd expect it to max out (on our
>>> quad-core test boxes) when you add client threads
>>>=20
>>> On Mon, Dec 7, 2009 at 10:38 AM, Timo Nentwig =
<timo.nentwig@wooga.net> wrote:
>>>> Hi!
>>>>=20
>>>> I just downloaded, installed, start cassandra and ran very simple =
"benchmark": write n times something with =
key=3D=3Dvalue=3D=3DtestInsertAndGetAndRemove_n (one thread).
>>>>=20
>>>> For n=3D=3D10 million on a 7200rpm HDD (4G RAM - there should have =
be "reasonably" free mem however I didn't check) this took 40min =
(insert()ing one after another). Reading them one by one in sequence =
delivers about 100/s, reading in 1.000er batches (i.e. multigetColumn()) =
takes 5-10s (depending on n, the higher the slower).
>>>>=20
>>>> Are this typical numbers for cassandra (0.5)? I actually took the =
configuration as it was.
>>=20
>>=20