Mailing-List: contact reviews-help@impala.incubator.apache.org; run by ezmlm
Precedence: bulk
Message-Id: <201710191818.v9JII8bP004760@ip-10-146-233-104.ec2.internal>
Date: Thu, 19 Oct 2017 18:18:07 +0000
From: "Tianyi Wang (Code Review)" <gerrit@cloudera.org>
To: Thomas Tauber-Marshall <tmarshall@cloudera.com>,
        Sailesh Mukil <sailesh@cloudera.com>,
        Tim Armstrong <tarmstrong@cloudera.com>, impala-cr@cloudera.com,
        reviews@impala.incubator.apache.org
Subject: =?UTF-8?Q?=5BImpala-ASF-CR=5D_IMPALA-5789=3A_Add_always_false_flag_in_bloom_filter=0A?=
In-Reply-To: <gerrit.1506648261000.If680240a3cd4583fc97c3192177d86d9567c4f8d@gerrit.cloudera.org>
References: <gerrit.1506648261000.If680240a3cd4583fc97c3192177d86d9567c4f8d@gerrit.cloudera.org>
Reply-To: twang@cloudera.com, impala-cr@cloudera.com, sailesh@cloudera.com,
        marcelk@gmail.com, tarmstrong@cloudera.com, tmarshall@cloudera.com,
        reviews@impala.incubator.apache.org
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Content-Disposition: inline
User-Agent: Gerrit/2.14.2
Content-Type: multipart/alternative; boundary="q/WwavllsKE="; charset=UTF-8
archived-at: Thu, 19 Oct 2017 18:18:14 -0000

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

Hello Thomas Tauber-Marshall, Sailesh Mukil, Tim Armstrong, 

I'd like you =
to reexamine a change=2E Please visit

    http://gerrit=2Ecloudera=2Eorg:8=
080/8170

to look at the new patch set (#7)=2E

Change subject: IMPALA-5789=
: Add always_false flag in bloom filter
=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=
=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=
=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=2E=
=2E=2E=2E=2E=2E=2E=2E=2E=2E

IMPALA-5789: Add always_false flag in bloom fi=
lter

This patch adds an always_false flag in bloom filters=2E The flag is =
set
if nothing has been inserted into the bloom filter=2E HdfsScanner uses
=
this flag to early terminate the scan at file and split granularities=2E

T=
esting: It passes existing tests=2E Two test cases are added checking
that =
an always-false runtime filter can filter out files and splits=2E
In single=
 node perf tests, time spent on primitive_empty_build_join_1
is reduced by =
75%=2E

Change-Id: If680240a3cd4583fc97c3192177d86d9567c4f8d
---
M be/src/c=
ommon/global-flags=2Ecc
M be/src/exec/base-sequence-scanner=2Ecc
M be/src/e=
xec/base-sequence-scanner=2Eh
M be/src/exec/filter-context=2Ecc
M be/src/ex=
ec/filter-context=2Eh
M be/src/exec/hdfs-parquet-scanner=2Ecc
M be/src/exec=
/hdfs-scan-node-base=2Ecc
M be/src/exec/hdfs-scan-node=2Ecc
M be/src/exec/h=
dfs-scanner=2Ecc
M be/src/runtime/coordinator-backend-state=2Ecc
M be/src/r=
untime/coordinator-backend-state=2Eh
M be/src/runtime/coordinator-filter-st=
ate=2Eh
M be/src/runtime/coordinator=2Ecc
M be/src/runtime/runtime-filter-i=
r=2Ecc
M be/src/runtime/runtime-filter=2Eh
M be/src/runtime/runtime-filter=
=2Einline=2Eh
M be/src/util/bloom-filter-ir=2Ecc
M be/src/util/bloom-filter=
=2Ecc
M be/src/util/bloom-filter=2Eh
M common/thrift/ImpalaInternalService=
=2Ethrift
A tests/custom_cluster/test_always_false_filter=2Epy
M tests/quer=
y_test/test_runtime_filters=2Epy
22 files changed, 200 insertions(+), 197 d=
eletions(-)


  git pull ssh://gerrit=2Ecloudera=2Eorg:29418/Impala-ASF ref=
s/changes/70/8170/7
-- 
To view, visit http://gerrit=2Ecloudera=2Eorg:8080/=
8170
To unsubscribe, visit http://gerrit=2Ecloudera=2Eorg:8080/settings

Ge=
rrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: newpatch=
set
Gerrit-Change-Id: If680240a3cd4583fc97c3192177d86d9567c4f8d
Gerrit-Chan=
ge-Number: 8170
Gerrit-PatchSet: 7
Gerrit-Owner: Tianyi Wang <twang@clouder=
a=2Ecom>
Gerrit-Reviewer: Sailesh Mukil <sailesh@cloudera=2Ecom>
Gerrit-Rev=
iewer: Thomas Tauber-Marshall <tmarshall@cloudera=2Ecom>
Gerrit-Reviewer: T=
ianyi Wang <twang@cloudera=2Ecom>
Gerrit-Reviewer: Tim Armstrong <tarmstron=
g@cloudera=2Ecom>

--q/WwavllsKE=--