Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Subject: Re: scylladb
To: user@cassandra.apache.org
References: <CAMwLmJZTF-yYFVi8WxGjU_gL38oUxL6d15Ub5QCoDj+i-cGPmw@mail.gmail.com>
 <CALcD3PvutGTcx9CmhH9PzLOR81ByJn0TwE3bfaLJq+sV-Oaiaw@mail.gmail.com>
 <CADVHTB8e7PGZxgLDWRRMnNx_JcD2PhUJDRcwz1RDK9Ap_km36Q@mail.gmail.com>
 <CAOUOv0HdM8TYJg-jBtep8CefEcC3G7Goy3FP1vBRaA0temRcrg@mail.gmail.com>
 <MWHPR2201MB1565623E6E532D7B79926A578C210@MWHPR2201MB1565.namprd22.prod.outlook.com>
 <CAEQiCCXXTdHX5NJc79bjz8+XOatHLjZ-V8Qq5+4m=Q2NU7E2GA@mail.gmail.com>
 <CADOsr-HtwraKpmRTj2eZ01Z87riFjEKPk5b2yRR5-eF3_Fa=7A@mail.gmail.com>
 <0c153096-af45-3a69-28ea-4a6471a47e04@scylladb.com>
 <CADQ6LY=nTRGGvvwY7Q+fvm7NjkyztAw1O7tmuxqT4tkRu188zA@mail.gmail.com>
 <7bd42a18-af9c-c9d5-98c9-083ced49cd5d@scylladb.com>
 <CADQ6LYmbBUOHWv3UgGvhZF25WEuPGtu6yuQ+sViT9QnsudQRyw@mail.gmail.com>
 <CAENxBwxyA2iwhgxykJftsp+07ZrigRHa_6WnFpq84=v9oC3C3Q@mail.gmail.com>
 <6d45df80-8193-73bc-0f1c-81079ab64d5c@scylladb.com>
 <CADOsr-HJ9ifkOB+4iOn2D66Re2TjENuj5pB79i2GZy3TrEc7mg@mail.gmail.com>
From: Avi Kivity <avi@scylladb.com>
Message-ID: <5f568408-9e60-1d09-64f9-60e4101f69a6@scylladb.com>
Date: Sat, 11 Mar 2017 22:51:35 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101
 Thunderbird/45.7.0
MIME-Version: 1.0
In-Reply-To: <CADOsr-HJ9ifkOB+4iOn2D66Re2TjENuj5pB79i2GZy3TrEc7mg@mail.gmail.com>
Content-Type: multipart/alternative;
 boundary="------------A6F2D8B8F62014B43AC543E8"
archived-at: Sat, 11 Mar 2017 20:52:20 -0000

This is a multi-part message in MIME format.
--------------A6F2D8B8F62014B43AC543E8
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit

Scylla uses a the seastar framework, which provides for both user-level 
thread scheduling and simple run-to-completion tasks.

Huge pages are limited to 2MB (and 1GB, but these aren't available as 
transparent hugepages).

On 03/11/2017 10:26 PM, Kant Kodali wrote:
> @Dor
>
> 1) You guys have a CPU scheduler? you mean user level thread Scheduler 
> that maps user level threads to kernel level threads? I thought C++ by 
> default creates native kernel threads but sure nothing will stop 
> someone to create a user level scheduling library if that's what you 
> are talking about?
> 2) How can one create THP of size 1KB? According to this post 
> <https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Performance_Tuning_Guide/s-memory-transhuge.html> it 
> looks like the valid values 2MB and 1GB.
>
> Thanks,
> kant
>
> On Sat, Mar 11, 2017 at 11:41 AM, Avi Kivity <avi@scylladb.com 
> <mailto:avi@scylladb.com>> wrote:
>
>     Agreed, I'd recommend to treat benchmarks as a rough guide to see
>     where there is potential, and follow through with your own tests.
>
>     On 03/11/2017 09:37 PM, Edward Capriolo wrote:
>>
>>     Benchmarks are great for FUDly blog posts. Real world work loads
>>     matter more. Every NoSQL vendor wins their benchmarks.
>
>
>


--------------A6F2D8B8F62014B43AC543E8
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: 8bit

<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Scylla uses a the seastar framework,
      which provides for both user-level thread scheduling and simple
      run-to-completion tasks.<br>
      <br>
      Huge pages are limited to 2MB (and 1GB, but these aren't available
      as transparent hugepages).<br>
      <br>
      On 03/11/2017 10:26 PM, Kant Kodali wrote:<br>
    </div>
    <blockquote
cite="mid:CADOsr-HJ9ifkOB+4iOn2D66Re2TjENuj5pB79i2GZy3TrEc7mg@mail.gmail.com"
      type="cite">
      <div dir="ltr">@Dor 
        <div><br>
        </div>
        <div>1) You guys have a CPU scheduler? you mean user level
          thread Scheduler that maps user level threads to kernel level
          threads? I thought C++ by default creates native kernel
          threads but sure nothing will stop someone to create a user
          level scheduling library if that's what you are talking about?</div>
        <div>2) How can one create THP of size 1KB? According to this <a
            moz-do-not-send="true"
href="https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Performance_Tuning_Guide/s-memory-transhuge.html">post</a> it
          looks like the valid values 2MB and 1GB.</div>
        <div><br>
        </div>
        <div>Thanks,</div>
        <div>kant</div>
      </div>
      <div class="gmail_extra"><br>
        <div class="gmail_quote">On Sat, Mar 11, 2017 at 11:41 AM, Avi
          Kivity <span dir="ltr">&lt;<a moz-do-not-send="true"
              href="mailto:avi@scylladb.com" target="_blank">avi@scylladb.com</a>&gt;</span>
          wrote:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0
            .8ex;border-left:1px #ccc solid;padding-left:1ex">
            <div bgcolor="#FFFFFF" text="#000000">
              <div class="m_6948238497389291502moz-cite-prefix">Agreed,
                I'd recommend to treat benchmarks as a rough guide to
                see where there is potential, and follow through with
                your own tests.<span class=""><br>
                  <br>
                  On 03/11/2017 09:37 PM, Edward Capriolo wrote:<br>
                </span></div>
              <span class="">
                <blockquote type="cite">
                  <div dir="ltr"><br>
                    <div class="gmail_extra">Benchmarks are great for
                      FUDly blog posts. Real world work loads matter
                      more. Every NoSQL vendor wins their benchmarks.</div>
                  </div>
                </blockquote>
                <p><br>
                </p>
              </span></div>
          </blockquote>
        </div>
        <br>
      </div>
    </blockquote>
    <br>
    <br>
    <p><br>
    </p>
  </body>
</html>

--------------A6F2D8B8F62014B43AC543E8--