Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Received-SPF: pass (athena.apache.org: domain of tivv00@gmail.com designates
 209.85.215.172 as permitted sender)
Message-ID: <4F071868.8020104@gmail.com>
Date: Fri, 06 Jan 2012 17:51:04 +0200
From: Vitalii Tymchyshyn <tivv00@gmail.com>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US;
 rv:1.9.2.24) Gecko/20111109 Lightning/1.0b2 Thunderbird/3.1.16
MIME-Version: 1.0
To: user@cassandra.apache.org
Subject: Re: Should I throttle deletes?
References: 
 <CAAQs3wtdMK6hY-ipRAzXpAxuX4FNiaeryJHv9hJ9wYp6D+ZeKQ@mail.gmail.com>
	<4F04C10E.1000904@bnl.gov>
	<A31E1470-678A-403B-B611-ECC87E7D7F9F@thelastpickle.com>
	<4F05F9C8.5050601@bnl.gov>
	<CAHwsXYmF+R_oBjLQQcwjWnZ74_054yJib7uL-VZF3GKdrtbmUg@mail.gmail.com>
	<4F070268.5030001@gmail.com>
 <CAHwsXYn4YyY0CuBTjtBrcvxRywmg-zEN7TQz+hJBey-sp+BRGg@mail.gmail.com>
In-Reply-To: 
 <CAHwsXYn4YyY0CuBTjtBrcvxRywmg-zEN7TQz+hJBey-sp+BRGg@mail.gmail.com>
Content-Type: multipart/alternative;
 boundary="------------080701040807000106090804"

This is a multi-part message in MIME format.
--------------080701040807000106090804
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Do you mean on writes? Yes, your timeouts must be so that your write 
batch could complete until timeout elapsed. But this will lower write 
load, so reads should not timeout.

Best regards, Vitalii Tymchyshym

06.01.12 17:37, Philippe написав(ла):
>
> But you will then get timeouts.
>
> Le 6 janv. 2012 15:17, "Vitalii Tymchyshyn" <tivv00@gmail.com 
> <mailto:tivv00@gmail.com>> a écrit :
>
>     05.01.12 22:29, Philippe написав(ла):
>>
>>         Then I do have a question, what do people generally use as
>>         the batch size?
>>
>>     I used to do batches from 500 to 2000 like you do.
>>     After investigating issues such as the one you've encountered
>>     I've moved to batches of 20 for writes and 256 for reads.
>>     Everything is a lot smoother : no more timeouts.
>>
>     I'd better reduce mutation thread pool with concurrent_writes
>     setting. This will lower server load no matter, how many clients
>     are sending batches, at the same time you still have good batching.
>
>     Best regards, Vitalii Tymchyshyn
>


--------------080701040807000106090804
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
  <head>
    <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#ffffff" text="#000000">
    Do you mean on writes? Yes, your timeouts must be so that your write
    batch could complete until timeout elapsed. But this will lower
    write load, so reads should not timeout.<br>
    <br>
    Best regards, Vitalii Tymchyshym<br>
    <br>
    06.01.12 17:37, Philippe написав(ла):
    <blockquote
cite="mid:CAHwsXYn4YyY0CuBTjtBrcvxRywmg-zEN7TQz+hJBey-sp+BRGg@mail.gmail.com"
      type="cite">
      <p>But you will then get timeouts.</p>
      <div class="gmail_quote">Le 6 janv. 2012 15:17, "Vitalii
        Tymchyshyn" &lt;<a moz-do-not-send="true"
          href="mailto:tivv00@gmail.com">tivv00@gmail.com</a>&gt; a
        écrit :<br type="attribution">
        <blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt
          0.8ex; border-left: 1px solid rgb(204, 204, 204);
          padding-left: 1ex;">
          <div bgcolor="#ffffff" text="#000000"> 05.01.12 22:29,
            Philippe написав(ла):
            <blockquote type="cite">
              <div class="gmail_quote">
                <blockquote class="gmail_quote" style="margin: 0pt 0pt
                  0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204);
                  padding-left: 1ex;">Then I do have a question, what do
                  people generally use as the batch size?<br>
                </blockquote>
                <div> I used to do batches from 500 to 2000 like you do.</div>
                <div>After investigating issues such as the one you've
                  encountered I've moved to batches of 20 for writes and
                  256 for reads. Everything is a lot smoother : no more
                  timeouts.</div>
                <br>
              </div>
            </blockquote>
            I'd better reduce mutation thread pool with
            concurrent_writes setting. This will lower server load no
            matter, how many clients are sending batches, at the same
            time you still have good batching.<br>
            <br>
            Best regards, Vitalii Tymchyshyn<br>
          </div>
        </blockquote>
      </div>
    </blockquote>
    <br>
  </body>
</html>

--------------080701040807000106090804--