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: <CAGUtCHr7-713o=WzJn=BBWz4pm=XLGx1_anR=t+_PtrRGK-3jg@mail.gmail.com>
References: <CAGUtCHr7-713o=WzJn=BBWz4pm=XLGx1_anR=t+_PtrRGK-3jg@mail.gmail.com>
From: Jeff Kubina <jeff.kubina@gmail.com>
Date: Wed, 2 Nov 2016 21:01:57 -0400
Message-ID: <CA+Vtps6JMs1qwQm1MT6WKBVPem6dbsrAgCvB9E2-rhVRv0MBzQ@mail.gmail.com>
Subject: Re: New Accumulo Blog Post
To: user@accumulo.apache.org
Content-Type: multipart/alternative; boundary=001a113949f0e5ec5c05405b1b0d
archived-at: Thu, 03 Nov 2016 01:02:23 -0000

--001a113949f0e5ec5c05405b1b0d
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Thanks for the blog post, very interesting read. Some questions ...

1. Are the operations "Writes mutation to tablet servers=E2=80=99 WAL/Sync =
or flush
tablet servers=E2=80=99 WAL" and "Adds mutations to sorted in memory map of=
 each
tablet." performed by threads in parallel?

2. Could the latency of hsync-ing the WALs be overcome by modifying
Accumulo to write them to a separate SSD-only HDFS? To maintain data
locality it would require two datanode processes (one for the HDDs and one
for the SSD), running on the same node, which is not hard to do.

--001a113949f0e5ec5c05405b1b0d
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Thanks for the blog post, very interesting read. Some ques=
tions ...<div><br></div><div>1. Are the operations &quot;Writes mutation to=
 tablet servers=E2=80=99 WAL/Sync or flush tablet servers=E2=80=99 WAL&quot=
; and &quot;Adds mutations to sorted in memory map of each tablet.&quot; pe=
rformed by threads in parallel?</div><div><br></div>2. Could the latency of=
 hsync-ing the WALs be overcome by modifying Accumulo to write them to a se=
parate SSD-only HDFS? To maintain data locality it would require two datano=
de processes (one for the HDDs and one for the SSD), running on the same no=
de, which is not hard to do.<div><br></div></div>

--001a113949f0e5ec5c05405b1b0d--