Mailing-List: contact user-help@accumulo.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@accumulo.apache.org
MIME-Version: 1.0
In-Reply-To: 
 <CALs3O6gd4MMi0neG7+_87BXVuwUDZDThSLy_WAecZCN0Ty8Gog@mail.gmail.com>
References: 
 <CA+Vtps7UeXFM5SusH06-vyVjMbbJpEK97q6LXZJdDMPMQhoRgA@mail.gmail.com>
 <CALs3O6gd4MMi0neG7+_87BXVuwUDZDThSLy_WAecZCN0Ty8Gog@mail.gmail.com>
From: Jeff Kubina <jeff.kubina@gmail.com>
Date: Mon, 17 Aug 2015 18:01:38 -0400
Message-ID: 
 <CA+Vtps7txzi4oWPeZvJM3FYUaPaVPRm--BdYjsOcBc-=dV7WxQ@mail.gmail.com>
Subject: Re: Question about configuring the linux niceness of tablet servers?
To: user@accumulo.apache.org
Content-Type: multipart/alternative; boundary=001a114082ea42b220051d88f386

--001a114082ea42b220051d88f386
Content-Type: text/plain; charset=UTF-8

More like a mapreduce task process.

-- 
Jeff Kubina
410-988-4436


On Mon, Aug 17, 2015 at 5:33 PM, William Slacum <wslacum@gmail.com> wrote:

> By "Hadoop" do you mean a Yarn NodeManager process?
>
> On Mon, Aug 17, 2015 at 4:21 PM, Jeff Kubina <jeff.kubina@gmail.com>
> wrote:
>
>> On each of the processing nodes in our cluster we have running 1) HDFS
>> (datanode), 2) Accumulo (tablet server), and 3) Hadoop. Since Accumulo
>> depends on the HDFS, and Hadoop depends on the HDFS and sometimes on
>> Accumulo, we are considering setting the niceness of HDFS to 0 (the current
>> value), Accumulo to 1, and Hadoop to 2 on each of the nodes. The objective
>> is to improve the real time performance of Accumulo.
>>
>> Does anyone have experience configuring their cluster in a similar manner
>> that they can share?  Are there any serious cons to doing this?
>>
>>
>

--001a114082ea42b220051d88f386
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">More like a mapreduce task process.<br></div><div class=3D=
"gmail_extra"><br clear=3D"all"><div><div class=3D"gmail_signature"><div di=
r=3D"ltr"><div>--=C2=A0</div>Jeff Kubina<br>410-988-4436<div><br></div></di=
v></div></div>
<br><div class=3D"gmail_quote">On Mon, Aug 17, 2015 at 5:33 PM, William Sla=
cum <span dir=3D"ltr">&lt;<a href=3D"mailto:wslacum@gmail.com" target=3D"_b=
lank">wslacum@gmail.com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail=
_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:=
1ex"><div dir=3D"ltr">By &quot;Hadoop&quot; do you mean a Yarn NodeManager =
process?<br></div><div class=3D"HOEnZb"><div class=3D"h5"><div class=3D"gma=
il_extra"><br><div class=3D"gmail_quote">On Mon, Aug 17, 2015 at 4:21 PM, J=
eff Kubina <span dir=3D"ltr">&lt;<a href=3D"mailto:jeff.kubina@gmail.com" t=
arget=3D"_blank">jeff.kubina@gmail.com</a>&gt;</span> wrote:<br><blockquote=
 class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc soli=
d;padding-left:1ex"><div dir=3D"ltr"><div>On each of the processing nodes i=
n our cluster we have running 1) HDFS (datanode),  2) Accumulo (tablet serv=
er), and 3) Hadoop. Since Accumulo depends on the HDFS, and Hadoop depends =
on the HDFS and sometimes on Accumulo, we are considering setting the nicen=
ess of HDFS to 0 (the current value), Accumulo to 1, and Hadoop to 2 on eac=
h of the nodes. The objective is to improve the real time performance of Ac=
cumulo.<br><br></div>Does anyone have experience configuring their cluster =
in a similar manner that they can share?=C2=A0 Are there any serious cons t=
o doing this?<br><br></div>
</blockquote></div><br></div>
</div></div></blockquote></div><br></div>

--001a114082ea42b220051d88f386--