You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@dubbo.apache.org by Zonghai Shang <yi...@apache.org> on 2018/09/12 06:32:11 UTC

[Dubbo benchmark ]The test results are mixed with the dubbo log.

Hi ,

I am using performance testing to find out that the output is confusing, it
would be better if the output is supported to the file.
dubbo-benchmark: https://github.com/dubbo/dubbo-benchmark
Here is my test output:

```
# Fork: 1 of 1
# Warmup Iteration   1: Sep 12, 2018 2:22:31 PM
org.springframework.context.support.ClassPathXmlApplicationContext
prepareRefresh
信息: Refreshing
org.springframework.context.support.ClassPathXmlApplicationContext@44dc6a41:
startup date [Wed Sep 12 14:22:31 CST 2018]; root of context hierarchy
Sep 12, 2018 2:22:31 PM
org.springframework.beans.factory.xml.XmlBeanDefinitionReader
loadBeanDefinitions
信息: Loading XML bean definitions from class path resource [consumer.xml]
14:22:31.775 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1] INFO
 c.a.d.common.logger.LoggerFactory - using logger:
com.alibaba.dubbo.common.logger.slf4j.Slf4jLoggerAdapter
14:22:32.040 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1] INFO
 c.a.d.r.protocol.dubbo.DubboProtocol -  [DUBBO] Optimizing the
serialization process for Kryo, FST, etc..., dubbo version: 2.6.1, current
host: 172.17.9.71
2.841 ±(99.9%) 0.015 ms/op
# Warmup Iteration   2: 2.464 ±(99.9%) 0.010 ms/op
# Warmup Iteration   3: 2.541 ±(99.9%) 0.007 ms/op
Iteration   1: 2.342 ±(99.9%) 0.005 ms/op
                 listUser·p0.00:   0.785 ms/op
```

Thanks
yiji

Re: [Dubbo benchmark ]The test results are mixed with the dubbo log.

Posted by Zonghai Shang <yi...@apache.org>.
>  I plan to support the -f option to output to a file

I have already completed this feature. pull request:
https://github.com/dubbo/dubbo-benchmark/commit/c22b29d4944778390642baa1e5da2380600c64b8

Best regards,
yiji

Ian Luo <ia...@gmail.com> 于2018年9月20日周四 下午12:40写道:

> We don't need to release the result. This project is for bottleneck
> analysis only.
>
> On Wed, Sep 19, 2018 at 7:29 AM Huxing Zhang <hu...@apache.org> wrote:
>
> > On Fri, Sep 14, 2018 at 10:25 AM Zonghai Shang <yi...@apache.org> wrote:
> > >
> > > > I plan to support the -f option to output to a file
> > >
> > > I will finish it this weekend.
> > >
> > > >  how to release the official benchmark data?
> > >
> > > I think the currently available benchmark test cases are incomplete and
> > > cannot be released at this time.
> >
> > I have't looked at the code yet. What do you think are needed to be
> > done before release the result?
> >
> > >
> > > Best regards,
> > > yiji
> > >
> > > jun liu <ke...@gmail.com> 于2018年9月14日周五 上午10:14写道:
> > >
> > > > > I plan to support the -f option to output to a file, by the way,
> this
> > > > > benchmark is great.
> > > >
> > > > I have checked the project, great job by Ian. Do we have a plan for
> > when
> > > > and how to release the official benchmark data?
> > > >
> > > > Best regards,
> > > > Jun
> > > >
> > > > > On 13 Sep 2018, at 16:24, Zonghai Shang <yi...@apache.org> wrote:
> > > > >
> > > > > I plan to support the -f option to output to a file, by the way,
> this
> > > > > benchmark is great.
> > > > >
> > > > > yiji
> > > > >
> > > > > Ian Luo <ia...@gmail.com> 于2018年9月12日周三 下午4:05写道:
> > > > >
> > > > >> Yiji,
> > > > >>
> > > > >> It doesn't support different format, and can be able to output
> into
> > a
> > > > file.
> > > > >> Pls. take a look at OptionsBuilder [1], feel free to enhance it.
> > > > >>
> > > > >> Regards,
> > > > >> -Ian.
> > > > >>
> > > > >> 1.
> > > > >>
> > > > >>
> > > >
> >
> https://github.com/dubbo/dubbo-benchmark/blob/master/client-base/src/main/java/org/apache/dubbo/benchmark/Client.java#L89
> > > > >>
> > > > >> On Wed, Sep 12, 2018 at 2:32 PM Zonghai Shang <yi...@apache.org>
> > wrote:
> > > > >>
> > > > >>> Hi ,
> > > > >>>
> > > > >>> I am using performance testing to find out that the output is
> > > > confusing,
> > > > >> it
> > > > >>> would be better if the output is supported to the file.
> > > > >>> dubbo-benchmark: https://github.com/dubbo/dubbo-benchmark
> > > > >>> Here is my test output:
> > > > >>>
> > > > >>> ```
> > > > >>> # Fork: 1 of 1
> > > > >>> # Warmup Iteration   1: Sep 12, 2018 2:22:31 PM
> > > > >>>
> org.springframework.context.support.ClassPathXmlApplicationContext
> > > > >>> prepareRefresh
> > > > >>> 信息: Refreshing
> > > > >>>
> > > > >>
> > > >
> >
> org.springframework.context.support.ClassPathXmlApplicationContext@44dc6a41
> > > > >>> :
> > > > >>> startup date [Wed Sep 12 14:22:31 CST 2018]; root of context
> > hierarchy
> > > > >>> Sep 12, 2018 2:22:31 PM
> > > > >>> org.springframework.beans.factory.xml.XmlBeanDefinitionReader
> > > > >>> loadBeanDefinitions
> > > > >>> 信息: Loading XML bean definitions from class path resource
> > > > [consumer.xml]
> > > > >>> 14:22:31.775
> > [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> > > > >> INFO
> > > > >>> c.a.d.common.logger.LoggerFactory - using logger:
> > > > >>> com.alibaba.dubbo.common.logger.slf4j.Slf4jLoggerAdapter
> > > > >>> 14:22:32.040
> > [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> > > > >> INFO
> > > > >>> c.a.d.r.protocol.dubbo.DubboProtocol -  [DUBBO] Optimizing the
> > > > >>> serialization process for Kryo, FST, etc..., dubbo version:
> 2.6.1,
> > > > >> current
> > > > >>> host: 172.17.9.71
> > > > >>> 2.841 ±(99.9%) 0.015 ms/op
> > > > >>> # Warmup Iteration   2: 2.464 ±(99.9%) 0.010 ms/op
> > > > >>> # Warmup Iteration   3: 2.541 ±(99.9%) 0.007 ms/op
> > > > >>> Iteration   1: 2.342 ±(99.9%) 0.005 ms/op
> > > > >>>                 listUser·p0.00:   0.785 ms/op
> > > > >>> ```
> > > > >>>
> > > > >>> Thanks
> > > > >>> yiji
> > > > >>>
> > > > >>
> > > >
> > > >
> >
> >
> >
> > --
> > Best Regards!
> > Huxing
> >
>

Re: [Dubbo benchmark ]The test results are mixed with the dubbo log.

Posted by Ian Luo <ia...@gmail.com>.
We don't need to release the result. This project is for bottleneck
analysis only.

On Wed, Sep 19, 2018 at 7:29 AM Huxing Zhang <hu...@apache.org> wrote:

> On Fri, Sep 14, 2018 at 10:25 AM Zonghai Shang <yi...@apache.org> wrote:
> >
> > > I plan to support the -f option to output to a file
> >
> > I will finish it this weekend.
> >
> > >  how to release the official benchmark data?
> >
> > I think the currently available benchmark test cases are incomplete and
> > cannot be released at this time.
>
> I have't looked at the code yet. What do you think are needed to be
> done before release the result?
>
> >
> > Best regards,
> > yiji
> >
> > jun liu <ke...@gmail.com> 于2018年9月14日周五 上午10:14写道:
> >
> > > > I plan to support the -f option to output to a file, by the way, this
> > > > benchmark is great.
> > >
> > > I have checked the project, great job by Ian. Do we have a plan for
> when
> > > and how to release the official benchmark data?
> > >
> > > Best regards,
> > > Jun
> > >
> > > > On 13 Sep 2018, at 16:24, Zonghai Shang <yi...@apache.org> wrote:
> > > >
> > > > I plan to support the -f option to output to a file, by the way, this
> > > > benchmark is great.
> > > >
> > > > yiji
> > > >
> > > > Ian Luo <ia...@gmail.com> 于2018年9月12日周三 下午4:05写道:
> > > >
> > > >> Yiji,
> > > >>
> > > >> It doesn't support different format, and can be able to output into
> a
> > > file.
> > > >> Pls. take a look at OptionsBuilder [1], feel free to enhance it.
> > > >>
> > > >> Regards,
> > > >> -Ian.
> > > >>
> > > >> 1.
> > > >>
> > > >>
> > >
> https://github.com/dubbo/dubbo-benchmark/blob/master/client-base/src/main/java/org/apache/dubbo/benchmark/Client.java#L89
> > > >>
> > > >> On Wed, Sep 12, 2018 at 2:32 PM Zonghai Shang <yi...@apache.org>
> wrote:
> > > >>
> > > >>> Hi ,
> > > >>>
> > > >>> I am using performance testing to find out that the output is
> > > confusing,
> > > >> it
> > > >>> would be better if the output is supported to the file.
> > > >>> dubbo-benchmark: https://github.com/dubbo/dubbo-benchmark
> > > >>> Here is my test output:
> > > >>>
> > > >>> ```
> > > >>> # Fork: 1 of 1
> > > >>> # Warmup Iteration   1: Sep 12, 2018 2:22:31 PM
> > > >>> org.springframework.context.support.ClassPathXmlApplicationContext
> > > >>> prepareRefresh
> > > >>> 信息: Refreshing
> > > >>>
> > > >>
> > >
> org.springframework.context.support.ClassPathXmlApplicationContext@44dc6a41
> > > >>> :
> > > >>> startup date [Wed Sep 12 14:22:31 CST 2018]; root of context
> hierarchy
> > > >>> Sep 12, 2018 2:22:31 PM
> > > >>> org.springframework.beans.factory.xml.XmlBeanDefinitionReader
> > > >>> loadBeanDefinitions
> > > >>> 信息: Loading XML bean definitions from class path resource
> > > [consumer.xml]
> > > >>> 14:22:31.775
> [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> > > >> INFO
> > > >>> c.a.d.common.logger.LoggerFactory - using logger:
> > > >>> com.alibaba.dubbo.common.logger.slf4j.Slf4jLoggerAdapter
> > > >>> 14:22:32.040
> [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> > > >> INFO
> > > >>> c.a.d.r.protocol.dubbo.DubboProtocol -  [DUBBO] Optimizing the
> > > >>> serialization process for Kryo, FST, etc..., dubbo version: 2.6.1,
> > > >> current
> > > >>> host: 172.17.9.71
> > > >>> 2.841 ±(99.9%) 0.015 ms/op
> > > >>> # Warmup Iteration   2: 2.464 ±(99.9%) 0.010 ms/op
> > > >>> # Warmup Iteration   3: 2.541 ±(99.9%) 0.007 ms/op
> > > >>> Iteration   1: 2.342 ±(99.9%) 0.005 ms/op
> > > >>>                 listUser·p0.00:   0.785 ms/op
> > > >>> ```
> > > >>>
> > > >>> Thanks
> > > >>> yiji
> > > >>>
> > > >>
> > >
> > >
>
>
>
> --
> Best Regards!
> Huxing
>

Re: [Dubbo benchmark ]The test results are mixed with the dubbo log.

Posted by Huxing Zhang <hu...@apache.org>.
On Fri, Sep 14, 2018 at 10:25 AM Zonghai Shang <yi...@apache.org> wrote:
>
> > I plan to support the -f option to output to a file
>
> I will finish it this weekend.
>
> >  how to release the official benchmark data?
>
> I think the currently available benchmark test cases are incomplete and
> cannot be released at this time.

I have't looked at the code yet. What do you think are needed to be
done before release the result?

>
> Best regards,
> yiji
>
> jun liu <ke...@gmail.com> 于2018年9月14日周五 上午10:14写道:
>
> > > I plan to support the -f option to output to a file, by the way, this
> > > benchmark is great.
> >
> > I have checked the project, great job by Ian. Do we have a plan for when
> > and how to release the official benchmark data?
> >
> > Best regards,
> > Jun
> >
> > > On 13 Sep 2018, at 16:24, Zonghai Shang <yi...@apache.org> wrote:
> > >
> > > I plan to support the -f option to output to a file, by the way, this
> > > benchmark is great.
> > >
> > > yiji
> > >
> > > Ian Luo <ia...@gmail.com> 于2018年9月12日周三 下午4:05写道:
> > >
> > >> Yiji,
> > >>
> > >> It doesn't support different format, and can be able to output into a
> > file.
> > >> Pls. take a look at OptionsBuilder [1], feel free to enhance it.
> > >>
> > >> Regards,
> > >> -Ian.
> > >>
> > >> 1.
> > >>
> > >>
> > https://github.com/dubbo/dubbo-benchmark/blob/master/client-base/src/main/java/org/apache/dubbo/benchmark/Client.java#L89
> > >>
> > >> On Wed, Sep 12, 2018 at 2:32 PM Zonghai Shang <yi...@apache.org> wrote:
> > >>
> > >>> Hi ,
> > >>>
> > >>> I am using performance testing to find out that the output is
> > confusing,
> > >> it
> > >>> would be better if the output is supported to the file.
> > >>> dubbo-benchmark: https://github.com/dubbo/dubbo-benchmark
> > >>> Here is my test output:
> > >>>
> > >>> ```
> > >>> # Fork: 1 of 1
> > >>> # Warmup Iteration   1: Sep 12, 2018 2:22:31 PM
> > >>> org.springframework.context.support.ClassPathXmlApplicationContext
> > >>> prepareRefresh
> > >>> 信息: Refreshing
> > >>>
> > >>
> > org.springframework.context.support.ClassPathXmlApplicationContext@44dc6a41
> > >>> :
> > >>> startup date [Wed Sep 12 14:22:31 CST 2018]; root of context hierarchy
> > >>> Sep 12, 2018 2:22:31 PM
> > >>> org.springframework.beans.factory.xml.XmlBeanDefinitionReader
> > >>> loadBeanDefinitions
> > >>> 信息: Loading XML bean definitions from class path resource
> > [consumer.xml]
> > >>> 14:22:31.775 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> > >> INFO
> > >>> c.a.d.common.logger.LoggerFactory - using logger:
> > >>> com.alibaba.dubbo.common.logger.slf4j.Slf4jLoggerAdapter
> > >>> 14:22:32.040 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> > >> INFO
> > >>> c.a.d.r.protocol.dubbo.DubboProtocol -  [DUBBO] Optimizing the
> > >>> serialization process for Kryo, FST, etc..., dubbo version: 2.6.1,
> > >> current
> > >>> host: 172.17.9.71
> > >>> 2.841 ±(99.9%) 0.015 ms/op
> > >>> # Warmup Iteration   2: 2.464 ±(99.9%) 0.010 ms/op
> > >>> # Warmup Iteration   3: 2.541 ±(99.9%) 0.007 ms/op
> > >>> Iteration   1: 2.342 ±(99.9%) 0.005 ms/op
> > >>>                 listUser·p0.00:   0.785 ms/op
> > >>> ```
> > >>>
> > >>> Thanks
> > >>> yiji
> > >>>
> > >>
> >
> >



-- 
Best Regards!
Huxing

Re: [Dubbo benchmark ]The test results are mixed with the dubbo log.

Posted by Zonghai Shang <yi...@apache.org>.
> I plan to support the -f option to output to a file

I will finish it this weekend.

>  how to release the official benchmark data?

I think the currently available benchmark test cases are incomplete and
cannot be released at this time.

Best regards,
yiji

jun liu <ke...@gmail.com> 于2018年9月14日周五 上午10:14写道:

> > I plan to support the -f option to output to a file, by the way, this
> > benchmark is great.
>
> I have checked the project, great job by Ian. Do we have a plan for when
> and how to release the official benchmark data?
>
> Best regards,
> Jun
>
> > On 13 Sep 2018, at 16:24, Zonghai Shang <yi...@apache.org> wrote:
> >
> > I plan to support the -f option to output to a file, by the way, this
> > benchmark is great.
> >
> > yiji
> >
> > Ian Luo <ia...@gmail.com> 于2018年9月12日周三 下午4:05写道:
> >
> >> Yiji,
> >>
> >> It doesn't support different format, and can be able to output into a
> file.
> >> Pls. take a look at OptionsBuilder [1], feel free to enhance it.
> >>
> >> Regards,
> >> -Ian.
> >>
> >> 1.
> >>
> >>
> https://github.com/dubbo/dubbo-benchmark/blob/master/client-base/src/main/java/org/apache/dubbo/benchmark/Client.java#L89
> >>
> >> On Wed, Sep 12, 2018 at 2:32 PM Zonghai Shang <yi...@apache.org> wrote:
> >>
> >>> Hi ,
> >>>
> >>> I am using performance testing to find out that the output is
> confusing,
> >> it
> >>> would be better if the output is supported to the file.
> >>> dubbo-benchmark: https://github.com/dubbo/dubbo-benchmark
> >>> Here is my test output:
> >>>
> >>> ```
> >>> # Fork: 1 of 1
> >>> # Warmup Iteration   1: Sep 12, 2018 2:22:31 PM
> >>> org.springframework.context.support.ClassPathXmlApplicationContext
> >>> prepareRefresh
> >>> 信息: Refreshing
> >>>
> >>
> org.springframework.context.support.ClassPathXmlApplicationContext@44dc6a41
> >>> :
> >>> startup date [Wed Sep 12 14:22:31 CST 2018]; root of context hierarchy
> >>> Sep 12, 2018 2:22:31 PM
> >>> org.springframework.beans.factory.xml.XmlBeanDefinitionReader
> >>> loadBeanDefinitions
> >>> 信息: Loading XML bean definitions from class path resource
> [consumer.xml]
> >>> 14:22:31.775 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> >> INFO
> >>> c.a.d.common.logger.LoggerFactory - using logger:
> >>> com.alibaba.dubbo.common.logger.slf4j.Slf4jLoggerAdapter
> >>> 14:22:32.040 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> >> INFO
> >>> c.a.d.r.protocol.dubbo.DubboProtocol -  [DUBBO] Optimizing the
> >>> serialization process for Kryo, FST, etc..., dubbo version: 2.6.1,
> >> current
> >>> host: 172.17.9.71
> >>> 2.841 ±(99.9%) 0.015 ms/op
> >>> # Warmup Iteration   2: 2.464 ±(99.9%) 0.010 ms/op
> >>> # Warmup Iteration   3: 2.541 ±(99.9%) 0.007 ms/op
> >>> Iteration   1: 2.342 ±(99.9%) 0.005 ms/op
> >>>                 listUser·p0.00:   0.785 ms/op
> >>> ```
> >>>
> >>> Thanks
> >>> yiji
> >>>
> >>
>
>

Re: [Dubbo benchmark ]The test results are mixed with the dubbo log.

Posted by jun liu <ke...@gmail.com>.
> I plan to support the -f option to output to a file, by the way, this
> benchmark is great.

I have checked the project, great job by Ian. Do we have a plan for when and how to release the official benchmark data? 

Best regards,
Jun

> On 13 Sep 2018, at 16:24, Zonghai Shang <yi...@apache.org> wrote:
> 
> I plan to support the -f option to output to a file, by the way, this
> benchmark is great.
> 
> yiji
> 
> Ian Luo <ia...@gmail.com> 于2018年9月12日周三 下午4:05写道:
> 
>> Yiji,
>> 
>> It doesn't support different format, and can be able to output into a file.
>> Pls. take a look at OptionsBuilder [1], feel free to enhance it.
>> 
>> Regards,
>> -Ian.
>> 
>> 1.
>> 
>> https://github.com/dubbo/dubbo-benchmark/blob/master/client-base/src/main/java/org/apache/dubbo/benchmark/Client.java#L89
>> 
>> On Wed, Sep 12, 2018 at 2:32 PM Zonghai Shang <yi...@apache.org> wrote:
>> 
>>> Hi ,
>>> 
>>> I am using performance testing to find out that the output is confusing,
>> it
>>> would be better if the output is supported to the file.
>>> dubbo-benchmark: https://github.com/dubbo/dubbo-benchmark
>>> Here is my test output:
>>> 
>>> ```
>>> # Fork: 1 of 1
>>> # Warmup Iteration   1: Sep 12, 2018 2:22:31 PM
>>> org.springframework.context.support.ClassPathXmlApplicationContext
>>> prepareRefresh
>>> 信息: Refreshing
>>> 
>> org.springframework.context.support.ClassPathXmlApplicationContext@44dc6a41
>>> :
>>> startup date [Wed Sep 12 14:22:31 CST 2018]; root of context hierarchy
>>> Sep 12, 2018 2:22:31 PM
>>> org.springframework.beans.factory.xml.XmlBeanDefinitionReader
>>> loadBeanDefinitions
>>> 信息: Loading XML bean definitions from class path resource [consumer.xml]
>>> 14:22:31.775 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
>> INFO
>>> c.a.d.common.logger.LoggerFactory - using logger:
>>> com.alibaba.dubbo.common.logger.slf4j.Slf4jLoggerAdapter
>>> 14:22:32.040 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
>> INFO
>>> c.a.d.r.protocol.dubbo.DubboProtocol -  [DUBBO] Optimizing the
>>> serialization process for Kryo, FST, etc..., dubbo version: 2.6.1,
>> current
>>> host: 172.17.9.71
>>> 2.841 ±(99.9%) 0.015 ms/op
>>> # Warmup Iteration   2: 2.464 ±(99.9%) 0.010 ms/op
>>> # Warmup Iteration   3: 2.541 ±(99.9%) 0.007 ms/op
>>> Iteration   1: 2.342 ±(99.9%) 0.005 ms/op
>>>                 listUser·p0.00:   0.785 ms/op
>>> ```
>>> 
>>> Thanks
>>> yiji
>>> 
>> 


Re: [Dubbo benchmark ]The test results are mixed with the dubbo log.

Posted by Zonghai Shang <yi...@apache.org>.
I plan to support the -f option to output to a file, by the way, this
benchmark is great.

yiji

Ian Luo <ia...@gmail.com> 于2018年9月12日周三 下午4:05写道:

> Yiji,
>
> It doesn't support different format, and can be able to output into a file.
> Pls. take a look at OptionsBuilder [1], feel free to enhance it.
>
> Regards,
> -Ian.
>
> 1.
>
> https://github.com/dubbo/dubbo-benchmark/blob/master/client-base/src/main/java/org/apache/dubbo/benchmark/Client.java#L89
>
> On Wed, Sep 12, 2018 at 2:32 PM Zonghai Shang <yi...@apache.org> wrote:
>
> > Hi ,
> >
> > I am using performance testing to find out that the output is confusing,
> it
> > would be better if the output is supported to the file.
> > dubbo-benchmark: https://github.com/dubbo/dubbo-benchmark
> > Here is my test output:
> >
> > ```
> > # Fork: 1 of 1
> > # Warmup Iteration   1: Sep 12, 2018 2:22:31 PM
> > org.springframework.context.support.ClassPathXmlApplicationContext
> > prepareRefresh
> > 信息: Refreshing
> >
> org.springframework.context.support.ClassPathXmlApplicationContext@44dc6a41
> > :
> > startup date [Wed Sep 12 14:22:31 CST 2018]; root of context hierarchy
> > Sep 12, 2018 2:22:31 PM
> > org.springframework.beans.factory.xml.XmlBeanDefinitionReader
> > loadBeanDefinitions
> > 信息: Loading XML bean definitions from class path resource [consumer.xml]
> > 14:22:31.775 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> INFO
> >  c.a.d.common.logger.LoggerFactory - using logger:
> > com.alibaba.dubbo.common.logger.slf4j.Slf4jLoggerAdapter
> > 14:22:32.040 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1]
> INFO
> >  c.a.d.r.protocol.dubbo.DubboProtocol -  [DUBBO] Optimizing the
> > serialization process for Kryo, FST, etc..., dubbo version: 2.6.1,
> current
> > host: 172.17.9.71
> > 2.841 ±(99.9%) 0.015 ms/op
> > # Warmup Iteration   2: 2.464 ±(99.9%) 0.010 ms/op
> > # Warmup Iteration   3: 2.541 ±(99.9%) 0.007 ms/op
> > Iteration   1: 2.342 ±(99.9%) 0.005 ms/op
> >                  listUser·p0.00:   0.785 ms/op
> > ```
> >
> > Thanks
> > yiji
> >
>

Re: [Dubbo benchmark ]The test results are mixed with the dubbo log.

Posted by Ian Luo <ia...@gmail.com>.
Yiji,

It doesn't support different format, and can be able to output into a file.
Pls. take a look at OptionsBuilder [1], feel free to enhance it.

Regards,
-Ian.

1.
https://github.com/dubbo/dubbo-benchmark/blob/master/client-base/src/main/java/org/apache/dubbo/benchmark/Client.java#L89

On Wed, Sep 12, 2018 at 2:32 PM Zonghai Shang <yi...@apache.org> wrote:

> Hi ,
>
> I am using performance testing to find out that the output is confusing, it
> would be better if the output is supported to the file.
> dubbo-benchmark: https://github.com/dubbo/dubbo-benchmark
> Here is my test output:
>
> ```
> # Fork: 1 of 1
> # Warmup Iteration   1: Sep 12, 2018 2:22:31 PM
> org.springframework.context.support.ClassPathXmlApplicationContext
> prepareRefresh
> 信息: Refreshing
> org.springframework.context.support.ClassPathXmlApplicationContext@44dc6a41
> :
> startup date [Wed Sep 12 14:22:31 CST 2018]; root of context hierarchy
> Sep 12, 2018 2:22:31 PM
> org.springframework.beans.factory.xml.XmlBeanDefinitionReader
> loadBeanDefinitions
> 信息: Loading XML bean definitions from class path resource [consumer.xml]
> 14:22:31.775 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1] INFO
>  c.a.d.common.logger.LoggerFactory - using logger:
> com.alibaba.dubbo.common.logger.slf4j.Slf4jLoggerAdapter
> 14:22:32.040 [org.apache.dubbo.benchmark.Client.listUser-jmh-worker-1] INFO
>  c.a.d.r.protocol.dubbo.DubboProtocol -  [DUBBO] Optimizing the
> serialization process for Kryo, FST, etc..., dubbo version: 2.6.1, current
> host: 172.17.9.71
> 2.841 ±(99.9%) 0.015 ms/op
> # Warmup Iteration   2: 2.464 ±(99.9%) 0.010 ms/op
> # Warmup Iteration   3: 2.541 ±(99.9%) 0.007 ms/op
> Iteration   1: 2.342 ±(99.9%) 0.005 ms/op
>                  listUser·p0.00:   0.785 ms/op
> ```
>
> Thanks
> yiji
>