Mailing-List: contact dev-help@impala.incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@impala.incubator.apache.org
Message-Id: <201608190253.u7J2rq4w011729@ip-10-146-233-104.ec2.internal>
Date: Fri, 19 Aug 2016 02:53:52 +0000
From: "Youwei Wang (Code Review)" <gerrit@cloudera.org>
To: Youwei Wang <youwei.a.wang@intel.com>, Jim Apple <jbapple@cloudera.com>,
        Tim Armstrong <tarmstrong@cloudera.com>, impala-cr@cloudera.com,
        dev@impala.incubator.apache.org
CC: Marcel Kornacker <marcel@cloudera.com>,
        Mostafa Mokhtar <mmokhtar@cloudera.com>,
        Alex Behm <alex.behm@cloudera.com>
Reply-To: youwei.a.wang@intel.com
Subject: =?UTF-8?Q?=5BImpala-CR=5D=28cdh5-trunk=29_IMPALA-2809=3A_Improve_ByteSwap_with_builtin_function_or_SSSE3_or_AVX2=2E=0A?=
In-Reply-To: <gerrit.1463366740000.I392ed5a8d5683f30f161282c228c1aedd7b648c1@gerrit.cloudera.org>
References: <gerrit.1463366740000.I392ed5a8d5683f30f161282c228c1aedd7b648c1@gerrit.cloudera.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Content-Disposition: inline
User-Agent: Gerrit/2.12.2
archived-at: Fri, 19 Aug 2016 02:53:59 -0000

Hello Jim Apple, Tim Armstrong,

I'd like you to reexamine a change.  Please visit

    http://gerrit.cloudera.org:8080/3081

to look at the new patch set (#41).

Change subject: IMPALA-2809: Improve ByteSwap with builtin function or SSSE3 or AVX2.
......................................................................

IMPALA-2809: Improve ByteSwap with builtin function or SSSE3 or AVX2.

Using SSSE3/AVX2 intrinsic to accelerate the function
"static inline void ByteSwap(void* dst, const void* src, int len)" of BitUtil class,
and a scalar byte-swap routine is added as fallback.
Also the runtime selector for CPUs of different capacity is included,
as well as performance test and data verification.
Brief performance comparison is listed here:
CPU: Intel(R) Core(TM) i5-4460  CPU@3.20GHz
Result:
I0725 20:47:02.402506  2078 bswap-benchmark.cc:117] Machine Info: Intel(R) Core(TM) i5-4460  CPU @ 3.20GHz
ByteSwap benchmark:        Function  iters/ms   10%ile   50%ile   90%ile     10%ile     50%ile     90%ile
                                                                         (relative) (relative) (relative)
---------------------------------------------------------------------------------------------------------
                         FastScalar                675      725      731         1X         1X         1X
                              SSSE3           6.12e+03  6.2e+03 6.23e+03      9.06X      8.55X      8.53X
                               AVX2           1.87e+04 1.88e+04 1.89e+04      27.7X      25.9X      25.9X
                               SIMD           1.82e+04 1.88e+04 1.89e+04        27X      25.9X      25.9X

Change-Id: I392ed5a8d5683f30f161282c228c1aedd7b648c1
---
M be/src/benchmarks/CMakeLists.txt
A be/src/benchmarks/bswap-benchmark.cc
M be/src/exprs/string-functions-ir.cc
M be/src/util/bit-util-test.cc
M be/src/util/bit-util.cc
M be/src/util/bit-util.h
6 files changed, 376 insertions(+), 51 deletions(-)


  git pull ssh://gerrit.cloudera.org:29418/Impala refs/changes/81/3081/41
-- 
To view, visit http://gerrit.cloudera.org:8080/3081
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I392ed5a8d5683f30f161282c228c1aedd7b648c1
Gerrit-PatchSet: 41
Gerrit-Project: Impala
Gerrit-Branch: cdh5-trunk
Gerrit-Owner: Youwei Wang <youwei.a.wang@intel.com>
Gerrit-Reviewer: Alex Behm <alex.behm@cloudera.com>
Gerrit-Reviewer: Jim Apple <jbapple@cloudera.com>
Gerrit-Reviewer: Marcel Kornacker <marcel@cloudera.com>
Gerrit-Reviewer: Mostafa Mokhtar <mmokhtar@cloudera.com>
Gerrit-Reviewer: Tim Armstrong <tarmstrong@cloudera.com>
Gerrit-Reviewer: Youwei Wang <youwei.a.wang@intel.com>