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 john@disqus.com designates
 209.85.210.171 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CAENxBwyB1kCDBocLDLcWHteBD7U2OwTVyN9gNDchSkDTL31oeA@mail.gmail.com>
References: 
 <CA+qt5VPMb4JTqxjMtrW+uWWJRrHO7N8fPaPSr=HSNSJjpKHUpA@mail.gmail.com>
	<CAENxBwyB1kCDBocLDLcWHteBD7U2OwTVyN9gNDchSkDTL31oeA@mail.gmail.com>
Date: Sun, 28 Apr 2013 14:49:58 -0700
Message-ID: 
 <CA+qt5VMn3Kc5aYrpps7qSCgjUiTWtwLhDP7H7aX6u+ZF8f_vng@mail.gmail.com>
Subject: Re: setcompactionthroughput and setstreamthroughput have no effect
From: John Watson <john@disqus.com>
To: user@cassandra.apache.org
Content-Type: multipart/alternative; boundary=90e6ba6e8a20df95a004db72bf51

--90e6ba6e8a20df95a004db72bf51
Content-Type: text/plain; charset=UTF-8

The help command says 0 to disable:
  setcompactionthroughput <value_in_mb> - Set the MB/s throughput cap for
compaction in the system, or 0 to disable throttling.
  setstreamthroughput  <value_in_mb> - Set the MB/s throughput cap for
streaming in the system, or 0 to disable throttling.

I also set both to 1000 and it also had no effect (just in case the
documentation was incorrect.)


On Sun, Apr 28, 2013 at 2:43 PM, Edward Capriolo <edlinuxguru@gmail.com>wrote:

> Out of curiosity. Why did you decide to set it to 0 rather then 99999.
> Does any documentation anywhere say that setting to 0 disables the feature?
> I have set streamthroughput higher and seen node join improvements. The
> features do work however they are probably not your limiting factor.
> Remember for stream you are setting Mega Bytes per second but network cards
> are measured in Mega Bits per second.
>
>
> On Sun, Apr 28, 2013 at 5:28 PM, John Watson <john@disqus.com> wrote:
>
>> Running these 2 commands are noop IO wise:
>>   nodetool setcompactionthroughput 0
>>   nodetool setstreamtrhoughput 0
>>
>> If trying to recover or rebuild nodes, it would be super helpful to get
>> more than ~120mbit/s of streaming throughput (per session or ~500mbit
>> total) and ~5% IO utilization in (8) 15k disk RAID10 (per cf).
>>
>> Even enabling multithreaded_compaction gives marginal improvements (1
>> additional thread doesn't help all that much and was only measurable in CPU
>> usage).
>>
>> I understand that these processes should take lower priority to servicing
>> reads and writes. However, in emergencies it would be a nice feature to
>> have a switch to recover a cluster ASAP.
>>
>> Thanks,
>>
>> John
>>
>
>

--90e6ba6e8a20df95a004db72bf51
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">The help command says 0 to disable:<div><div>=C2=A0 setcom=
pactionthroughput &lt;value_in_mb&gt; - Set the MB/s throughput cap for com=
paction in the system, or 0 to disable throttling.</div><div>=C2=A0 setstre=
amthroughput =C2=A0&lt;value_in_mb&gt; - Set the MB/s throughput cap for st=
reaming in the system, or 0 to disable throttling.</div>
</div><div><br></div><div style>I also set both to 1000 and it also had no =
effect (just in case the documentation was incorrect.)</div><div style><br>=
</div></div><div class=3D"gmail_extra"><br><br><div class=3D"gmail_quote">O=
n Sun, Apr 28, 2013 at 2:43 PM, Edward Capriolo <span dir=3D"ltr">&lt;<a hr=
ef=3D"mailto:edlinuxguru@gmail.com" target=3D"_blank">edlinuxguru@gmail.com=
</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div dir=3D"ltr">Out of curiosity. Why did y=
ou decide to set it to 0 rather then 99999. Does any documentation anywhere=
 say that setting to 0 disables the feature? I have set streamthroughput hi=
gher and seen node join improvements. The features do work however they are=
 probably not your limiting factor. Remember for stream you are setting Meg=
a Bytes per second but network cards are measured in Mega Bits per second.<=
br>

</div><div class=3D"HOEnZb"><div class=3D"h5"><div class=3D"gmail_extra"><b=
r><br><div class=3D"gmail_quote">On Sun, Apr 28, 2013 at 5:28 PM, John Wats=
on <span dir=3D"ltr">&lt;<a href=3D"mailto:john@disqus.com" target=3D"_blan=
k">john@disqus.com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
<div dir=3D"ltr">Running these 2 commands are noop IO wise:<div>=C2=A0 node=
tool setcompactionthroughput 0</div><div>=C2=A0 nodetool setstreamtrhoughpu=
t 0</div><div><br></div><div>If trying to recover or rebuild nodes, it woul=
d be super helpful to get more than ~120mbit/s of streaming throughput (per=
 session or ~500mbit total) and ~5% IO utilization in (8) 15k disk RAID10 (=
per cf).</div>


<div><br></div><div>Even enabling=C2=A0multithreaded_compaction gives margi=
nal improvements (1 additional thread doesn&#39;t help all that much and wa=
s only measurable in CPU usage).</div><div><br></div><div>
I understand that these processes should take lower priority to servicing r=
eads and writes. However, in emergencies it would be a nice feature to have=
 a switch to recover a cluster ASAP.</div><div><br></div><div>
Thanks,</div><div><br></div><div>John</div></div>
</blockquote></div><br></div>
</div></div></blockquote></div><br></div>

--90e6ba6e8a20df95a004db72bf51--