You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-dev@axis.apache.org by ax...@ws.apache.org on 2004/10/14 09:38:51 UTC
[jira] Created: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Message:
A new issue has been created in JIRA.
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Thu, 14 Oct 2004 12:38 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by "Davanum Srinivas (JIRA)" <ax...@ws.apache.org>.
[ http://nagoya.apache.org/jira/browse/AXIS-1608?page=comments#action_55112 ]
Davanum Srinivas commented on AXIS-1608:
----------------------------------------
This is MY zip file - http://nagoya.apache.org/jira/secure/attachment/18086/axisperftest.zip
which contains the modified version of the test case i got from venkat.
> Test case fro comparing Axis 1.2 against Axis 1.1
> -------------------------------------------------
>
> Key: AXIS-1608
> URL: http://nagoya.apache.org/jira/browse/AXIS-1608
> Project: Axis
> Type: Test
> Components: Basic Architecture
> Versions: 1.2RC2
> Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> Reporter: Venkat Reddy
> Attachments: axisperftest.zip, axisperftest.zip, perftest.cs, perftest2.zip, perftest2.zip
>
> My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> Timings (Average over 10 samples):
> ------------------------
> Axis 1.1 : 3891msecs
> Axis 1.2 : 5242 msec
> -- Venkat
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://nagoya.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by "Davanum Srinivas (JIRA)" <ax...@ws.apache.org>.
[ http://nagoya.apache.org/jira/browse/AXIS-1608?page=comments#action_55111 ]
Davanum Srinivas commented on AXIS-1608:
----------------------------------------
I don't see much difference between 1.1 and 1.2...Here's are the #'s i got. I ran the tests 3 times with 1 boot and 10 loops of calls from the client. See enclosed console.log for the transcript. See enclosed Perf.xls and Perf.htm for the numbers. All required bat files are enclosed. No, i did not use tomcat, i used SimpleAxisServer (started it from the bat file).
(Average Excluding Boot)
Average 1.1 1395.33
Average 1.2 1375.97
(Average Including Boot)
Average 1.1 1483.94
Average 1.2 1494.21
Thanks,
dims
> Test case fro comparing Axis 1.2 against Axis 1.1
> -------------------------------------------------
>
> Key: AXIS-1608
> URL: http://nagoya.apache.org/jira/browse/AXIS-1608
> Project: Axis
> Type: Test
> Components: Basic Architecture
> Versions: 1.2RC2
> Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> Reporter: Venkat Reddy
> Attachments: axisperftest.zip, axisperftest.zip, perftest.cs, perftest2.zip, perftest2.zip
>
> My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> Timings (Average over 10 samples):
> ------------------------
> Axis 1.1 : 3891msecs
> Axis 1.2 : 5242 msec
> -- Venkat
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://nagoya.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Davanum Srinivas
Created: Thu, 14 Oct 2004 6:55 AM
Body:
Venkat,
Can you please see if you can do this? Use .NET Framework SP1 as the client and compare the 2 axis versions on the server-side. Similarly compare the clients with .NET on the server. Then we can do some profiler stuff on the worst case scenario.
thanks,
dims
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54043
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Thu, 14 Oct 2004 6:55 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
Re: [jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by Davanum Srinivas <da...@gmail.com>.
edit client-config.wsdd
On Fri, 15 Oct 2004 17:20:13 +0530, Venkat Reddy <vr...@gmail.com> wrote:
> hi Dave,
>
> I do have the commons-httpclient-3.0-alpha1.jar in the axis/lib folder
> while i ran ant build on latest sources. Is there a way to check if
> the Axis is using CommonsHTTPSender?
>
> Thanks
> Venkat
>
>
>
>
> On Fri, 15 Oct 2004 07:41:26 -0400, David D. Lucas <dd...@lse.com> wrote:
> > The last I saw, CommonsHTTPSender was not the default, but I think it
> > should be. It adds caching of connections that play a big role,
> > especially when using HTTPS.
> >
> > Try it by rebuilding the axis.jar from source and include the Commons
> > HTTP jar in the lib. Then modify the client-config.wsdd file in the
> > source tree to use the Commons HTTP handler.
> >
> > Later,
> > Dave
> >
> >
> >
> > axis-dev@ws.apache.org wrote:
> > > The following comment has been added to this issue:
> > >
> > > Author: Venkat Reddy
> > > Created: Fri, 15 Oct 2004 3:04 AM
> > > Body:
> > > Following are the numbers with .Net Framework SP1 as client and Axis as server.
> > >
> > > Axis1.1: 3568
> > > Axis1.2: 5586 (1.57 times of v1.1)
> > >
> > > I'll do the opposite of it tomorrow (Axis on client, .Net on server)
> > >
> > > It appears to me that v1.2 is 1.5 times slower on the server and 1.3 times slower on client, and 1.9 times slower when used on both sides, according to my results.
> > >
> > > Regarding the CommonsHTTPSender, as you can see, i haven't specified any transport handler, and just using the axis jars out of the box. Isn't the CommonsHTTPSender default handler for transport? I'll look into this also tomorrow.
> > >
> > > -- Venkat
> > > ---------------------------------------------------------------------
> > > View this comment:
> > > http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54148
> > >
> > > ---------------------------------------------------------------------
> > > View the issue:
> > > http://issues.apache.org/jira/browse/AXIS-1608
> > >
> > > Here is an overview of the issue:
> > > ---------------------------------------------------------------------
> > > Key: AXIS-1608
> > > Summary: Test case fro comparing Axis 1.2 against Axis 1.1
> > > Type: Test
> > >
> > > Status: Unassigned
> > > Priority: Major
> > >
> > > Project: Axis
> > > Components:
> > > Basic Architecture
> > > Versions:
> > > 1.2RC2
> > >
> > > Assignee:
> > > Reporter: Venkat Reddy
> > >
> > > Created: Thu, 14 Oct 2004 12:38 AM
> > > Updated: Fri, 15 Oct 2004 3:04 AM
> > > Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> > >
> > > Description:
> > > My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> > >
> > > Timings (Average over 10 samples):
> > > ------------------------
> > > Axis 1.1 : 3891msecs
> > > Axis 1.2 : 5242 msec
> > >
> > >
> > > -- Venkat
> > >
> > >
> > >
> > > ---------------------------------------------------------------------
> > > JIRA INFORMATION:
> > > This message is automatically generated by JIRA.
> > >
> > > If you think it was sent incorrectly contact one of the administrators:
> > > http://issues.apache.org/jira/secure/Administrators.jspa
> > >
> > > If you want more information on JIRA, or have a bug to report see:
> > > http://www.atlassian.com/software/jira
> >
> >
> > --
> >
> > +------------------------------------------------------------+
> > | David Lucas mailto: ddlucas @ lse.com |
> > | Lucas Software Engineering, Inc. (740) 964-6248 Voice |
> > | Unix,Java,C++,CORBA,XML,EJB (614) 668-4020 Mobile |
> > | Middleware,Frameworks (888) 866-4728 Fax/Msg |
> > +------------------------------------------------------------+
> > | GPS Location: 40.0150 deg Lat, -82.6378 deg Long |
> > | IMHC: "Jesus Christ is the way, the truth, and the life." |
> > | IMHC: "I know where I am; I know where I'm going." <>< |
> > +------------------------------------------------------------+
> >
> > Notes: PGP Key Block=http://www.lse.com/~ddlucas/pgpblock.txt
> > IMHO="in my humble opinion" IMHC="in my humble conviction"
> > All trademarks above are those of their respective owners.
> >
> >
>
--
Davanum Srinivas - http://webservices.apache.org/~dims/
Re: [jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by Venkat Reddy <vr...@gmail.com>.
hi Dave,
I do have the commons-httpclient-3.0-alpha1.jar in the axis/lib folder
while i ran ant build on latest sources. Is there a way to check if
the Axis is using CommonsHTTPSender?
Thanks
Venkat
On Fri, 15 Oct 2004 07:41:26 -0400, David D. Lucas <dd...@lse.com> wrote:
> The last I saw, CommonsHTTPSender was not the default, but I think it
> should be. It adds caching of connections that play a big role,
> especially when using HTTPS.
>
> Try it by rebuilding the axis.jar from source and include the Commons
> HTTP jar in the lib. Then modify the client-config.wsdd file in the
> source tree to use the Commons HTTP handler.
>
> Later,
> Dave
>
>
>
> axis-dev@ws.apache.org wrote:
> > The following comment has been added to this issue:
> >
> > Author: Venkat Reddy
> > Created: Fri, 15 Oct 2004 3:04 AM
> > Body:
> > Following are the numbers with .Net Framework SP1 as client and Axis as server.
> >
> > Axis1.1: 3568
> > Axis1.2: 5586 (1.57 times of v1.1)
> >
> > I'll do the opposite of it tomorrow (Axis on client, .Net on server)
> >
> > It appears to me that v1.2 is 1.5 times slower on the server and 1.3 times slower on client, and 1.9 times slower when used on both sides, according to my results.
> >
> > Regarding the CommonsHTTPSender, as you can see, i haven't specified any transport handler, and just using the axis jars out of the box. Isn't the CommonsHTTPSender default handler for transport? I'll look into this also tomorrow.
> >
> > -- Venkat
> > ---------------------------------------------------------------------
> > View this comment:
> > http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54148
> >
> > ---------------------------------------------------------------------
> > View the issue:
> > http://issues.apache.org/jira/browse/AXIS-1608
> >
> > Here is an overview of the issue:
> > ---------------------------------------------------------------------
> > Key: AXIS-1608
> > Summary: Test case fro comparing Axis 1.2 against Axis 1.1
> > Type: Test
> >
> > Status: Unassigned
> > Priority: Major
> >
> > Project: Axis
> > Components:
> > Basic Architecture
> > Versions:
> > 1.2RC2
> >
> > Assignee:
> > Reporter: Venkat Reddy
> >
> > Created: Thu, 14 Oct 2004 12:38 AM
> > Updated: Fri, 15 Oct 2004 3:04 AM
> > Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> >
> > Description:
> > My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> >
> > Timings (Average over 10 samples):
> > ------------------------
> > Axis 1.1 : 3891msecs
> > Axis 1.2 : 5242 msec
> >
> >
> > -- Venkat
> >
> >
> >
> > ---------------------------------------------------------------------
> > JIRA INFORMATION:
> > This message is automatically generated by JIRA.
> >
> > If you think it was sent incorrectly contact one of the administrators:
> > http://issues.apache.org/jira/secure/Administrators.jspa
> >
> > If you want more information on JIRA, or have a bug to report see:
> > http://www.atlassian.com/software/jira
>
>
> --
>
> +------------------------------------------------------------+
> | David Lucas mailto: ddlucas @ lse.com |
> | Lucas Software Engineering, Inc. (740) 964-6248 Voice |
> | Unix,Java,C++,CORBA,XML,EJB (614) 668-4020 Mobile |
> | Middleware,Frameworks (888) 866-4728 Fax/Msg |
> +------------------------------------------------------------+
> | GPS Location: 40.0150 deg Lat, -82.6378 deg Long |
> | IMHC: "Jesus Christ is the way, the truth, and the life." |
> | IMHC: "I know where I am; I know where I'm going." <>< |
> +------------------------------------------------------------+
>
> Notes: PGP Key Block=http://www.lse.com/~ddlucas/pgpblock.txt
> IMHO="in my humble opinion" IMHC="in my humble conviction"
> All trademarks above are those of their respective owners.
>
>
Re: [jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2
against Axis 1.1
Posted by "David D. Lucas" <dd...@lse.com>.
The last I saw, CommonsHTTPSender was not the default, but I think it
should be. It adds caching of connections that play a big role,
especially when using HTTPS.
Try it by rebuilding the axis.jar from source and include the Commons
HTTP jar in the lib. Then modify the client-config.wsdd file in the
source tree to use the Commons HTTP handler.
Later,
Dave
axis-dev@ws.apache.org wrote:
> The following comment has been added to this issue:
>
> Author: Venkat Reddy
> Created: Fri, 15 Oct 2004 3:04 AM
> Body:
> Following are the numbers with .Net Framework SP1 as client and Axis as server.
>
> Axis1.1: 3568
> Axis1.2: 5586 (1.57 times of v1.1)
>
> I'll do the opposite of it tomorrow (Axis on client, .Net on server)
>
> It appears to me that v1.2 is 1.5 times slower on the server and 1.3 times slower on client, and 1.9 times slower when used on both sides, according to my results.
>
> Regarding the CommonsHTTPSender, as you can see, i haven't specified any transport handler, and just using the axis jars out of the box. Isn't the CommonsHTTPSender default handler for transport? I'll look into this also tomorrow.
>
> -- Venkat
> ---------------------------------------------------------------------
> View this comment:
> http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54148
>
> ---------------------------------------------------------------------
> View the issue:
> http://issues.apache.org/jira/browse/AXIS-1608
>
> Here is an overview of the issue:
> ---------------------------------------------------------------------
> Key: AXIS-1608
> Summary: Test case fro comparing Axis 1.2 against Axis 1.1
> Type: Test
>
> Status: Unassigned
> Priority: Major
>
> Project: Axis
> Components:
> Basic Architecture
> Versions:
> 1.2RC2
>
> Assignee:
> Reporter: Venkat Reddy
>
> Created: Thu, 14 Oct 2004 12:38 AM
> Updated: Fri, 15 Oct 2004 3:04 AM
> Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
>
> Description:
> My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
>
> Timings (Average over 10 samples):
> ------------------------
> Axis 1.1 : 3891msecs
> Axis 1.2 : 5242 msec
>
>
> -- Venkat
>
>
>
> ---------------------------------------------------------------------
> JIRA INFORMATION:
> This message is automatically generated by JIRA.
>
> If you think it was sent incorrectly contact one of the administrators:
> http://issues.apache.org/jira/secure/Administrators.jspa
>
> If you want more information on JIRA, or have a bug to report see:
> http://www.atlassian.com/software/jira
--
+------------------------------------------------------------+
| David Lucas mailto: ddlucas @ lse.com |
| Lucas Software Engineering, Inc. (740) 964-6248 Voice |
| Unix,Java,C++,CORBA,XML,EJB (614) 668-4020 Mobile |
| Middleware,Frameworks (888) 866-4728 Fax/Msg |
+------------------------------------------------------------+
| GPS Location: 40.0150 deg Lat, -82.6378 deg Long |
| IMHC: "Jesus Christ is the way, the truth, and the life." |
| IMHC: "I know where I am; I know where I'm going." <>< |
+------------------------------------------------------------+
Notes: PGP Key Block=http://www.lse.com/~ddlucas/pgpblock.txt
IMHO="in my humble opinion" IMHC="in my humble conviction"
All trademarks above are those of their respective owners.
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Venkat Reddy
Created: Fri, 15 Oct 2004 3:04 AM
Body:
Following are the numbers with .Net Framework SP1 as client and Axis as server.
Axis1.1: 3568
Axis1.2: 5586 (1.57 times of v1.1)
I'll do the opposite of it tomorrow (Axis on client, .Net on server)
It appears to me that v1.2 is 1.5 times slower on the server and 1.3 times slower on client, and 1.9 times slower when used on both sides, according to my results.
Regarding the CommonsHTTPSender, as you can see, i haven't specified any transport handler, and just using the axis jars out of the box. Isn't the CommonsHTTPSender default handler for transport? I'll look into this also tomorrow.
-- Venkat
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54148
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Fri, 15 Oct 2004 3:04 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Davanum Srinivas
Created: Sun, 17 Oct 2004 3:45 PM
Body:
Venkat,
Please do a bootstrap call to initialize both the server and the client BEFORE starting the timing.
CXObjectType boot = service.getObject("4");
Date start = new Date();
for(int i=0; i<100; ++i)
{
CXObjectType cxobj = service.getObject("4");
}
Date end = new Date();
-- dims
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54275
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Sun, 17 Oct 2004 3:45 PM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by "Eric Hansen (JIRA)" <ax...@ws.apache.org>.
[ http://nagoya.apache.org/jira/browse/AXIS-1608?page=history ]
Eric Hansen updated AXIS-1608:
------------------------------
Attachment: patch
The deserialization process was found under profiling to be heavily impacted by the creation of Text nodes. Those nodes are created using the owner document on the MessageElement. That class' method for finding the owner document implemented 3 strategies, the preferred one of which was particularly inefficient. Specifically, that implementation walked up the tree of nodes all the way to the root document, making it most noticeable in deeply nested documents. In our tests, a handful of message deserializations resulted in over 6 million calls to walk up the node tree.
The second strategy instead uses the DeserializationContext to locate the owner document with only two method calls. The attached patch reorders these strategies so that this is the preferred approach, with the one described above being used as a fallback for cases in which no deserialization context is present. Based on analysis of the source code, however, it wasn't clear that any such cases could actually arise.
> Test case fro comparing Axis 1.2 against Axis 1.1
> -------------------------------------------------
>
> Key: AXIS-1608
> URL: http://nagoya.apache.org/jira/browse/AXIS-1608
> Project: Axis
> Type: Test
> Components: Basic Architecture
> Versions: 1.2RC2
> Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> Reporter: Venkat Reddy
> Attachments: Perf.htm, axisperftest.zip, axisperftest.zip, patch, perftest.cs, perftest2.zip, perftest2.zip
>
> My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> Timings (Average over 10 samples):
> ------------------------
> Axis 1.1 : 3891msecs
> Axis 1.2 : 5242 msec
> -- Venkat
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://nagoya.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following issue has been updated:
Updater: Venkat Reddy (mailto:vreddyp@gmail.com)
Date: Fri, 15 Oct 2004 3:07 AM
Comment:
perftest.cs: CSharp Client for test
Changes:
Attachment changed to perftest.cs
---------------------------------------------------------------------
For a full history of the issue, see:
http://issues.apache.org/jira/browse/AXIS-1608?page=history
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Fri, 15 Oct 2004 3:07 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following issue has been updated:
Updater: Venkat Reddy (mailto:vreddyp@gmail.com)
Date: Thu, 14 Oct 2004 12:39 AM
Comment:
Test case attached.
Changes:
Attachment changed to perftest2.zip
---------------------------------------------------------------------
For a full history of the issue, see:
http://issues.apache.org/jira/browse/AXIS-1608?page=history
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Thu, 14 Oct 2004 12:39 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Venkat Reddy
Created: Thu, 14 Oct 2004 1:49 AM
Body:
Correction to my earlier timings:
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891 msec
Axis 1.2 : 7445 msec (almost twice as much 1.1)
The earlier observations of Axis1.2 being 1.35 times slower was due to using v1.2 on client side and v1.1 on server-side (though unintentionally). The latest observatiion of "twice slower" referes to using Axis1.2 on both client and server sides.
-- Venkat
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54036
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Thu, 14 Oct 2004 1:49 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Resolved: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by "Davanum Srinivas (JIRA)" <ax...@ws.apache.org>.
[ http://nagoya.apache.org/jira/browse/AXIS-1608?page=history ]
Davanum Srinivas resolved AXIS-1608:
------------------------------------
Resolution: Fixed
Closing Bug as resolved.
thanks,
dims
> Test case fro comparing Axis 1.2 against Axis 1.1
> -------------------------------------------------
>
> Key: AXIS-1608
> URL: http://nagoya.apache.org/jira/browse/AXIS-1608
> Project: Axis
> Type: Test
> Components: Basic Architecture
> Versions: 1.2RC2
> Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> Reporter: Venkat Reddy
> Attachments: Perf.htm, axisperftest.zip, axisperftest.zip, patch, perftest.cs, perftest2.zip, perftest2.zip
>
> My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> Timings (Average over 10 samples):
> ------------------------
> Axis 1.1 : 3891msecs
> Axis 1.2 : 5242 msec
> -- Venkat
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://nagoya.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following issue has been updated:
Updater: Venkat Reddy (mailto:vreddyp@gmail.com)
Date: Sun, 17 Oct 2004 10:51 PM
Comment:
Sources used for the tests.
The full test logs are available in axis.log.
Changes:
Attachment changed to perftest2.zip
---------------------------------------------------------------------
For a full history of the issue, see:
http://issues.apache.org/jira/browse/AXIS-1608?page=history
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Sun, 17 Oct 2004 10:51 PM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Dan Ciarniello
Created: Mon, 18 Oct 2004 9:17 AM
Body:
I've re-run my tests with a bootstrap call as well. The results are (average of 5 runs for each test)
Axis 1.1: 1921 ms
Axis 1.2: 4664 ms
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54311
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Mon, 18 Oct 2004 9:17 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by "Davanum Srinivas (JIRA)" <ax...@ws.apache.org>.
[ http://nagoya.apache.org/jira/browse/AXIS-1608?page=history ]
Davanum Srinivas updated AXIS-1608:
-----------------------------------
Attachment: Perf.htm
Standalone html with perf #'s
> Test case fro comparing Axis 1.2 against Axis 1.1
> -------------------------------------------------
>
> Key: AXIS-1608
> URL: http://nagoya.apache.org/jira/browse/AXIS-1608
> Project: Axis
> Type: Test
> Components: Basic Architecture
> Versions: 1.2RC2
> Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> Reporter: Venkat Reddy
> Attachments: Perf.htm, axisperftest.zip, axisperftest.zip, perftest.cs, perftest2.zip, perftest2.zip
>
> My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> Timings (Average over 10 samples):
> ------------------------
> Axis 1.1 : 3891msecs
> Axis 1.2 : 5242 msec
> -- Venkat
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://nagoya.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Venkat Reddy
Created: Sun, 17 Oct 2004 10:21 PM
Body:
In any case, i have used same test conditions and procedure for both v1.1 and v1.2. That means both versions are tested with simple usage without the bootstrap call. My humble opinion is that we better start identifying the bottlenecks in v1.2, instead of further tests for good results.
Did anyone try and produce better results?
-- Venkat
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54280
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Sun, 17 Oct 2004 10:21 PM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by "Davanum Srinivas (JIRA)" <ax...@ws.apache.org>.
[ http://nagoya.apache.org/jira/browse/AXIS-1608?page=comments#action_55114 ]
Davanum Srinivas commented on AXIS-1608:
----------------------------------------
For those lazy folks who don't want to unzip :)
http://nagoya.apache.org/jira/secure/attachment/18087/Perf.htm
-- dims
> Test case fro comparing Axis 1.2 against Axis 1.1
> -------------------------------------------------
>
> Key: AXIS-1608
> URL: http://nagoya.apache.org/jira/browse/AXIS-1608
> Project: Axis
> Type: Test
> Components: Basic Architecture
> Versions: 1.2RC2
> Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> Reporter: Venkat Reddy
> Attachments: Perf.htm, axisperftest.zip, axisperftest.zip, perftest.cs, perftest2.zip, perftest2.zip
>
> My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> Timings (Average over 10 samples):
> ------------------------
> Axis 1.1 : 3891msecs
> Axis 1.2 : 5242 msec
> -- Venkat
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://nagoya.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by "Davanum Srinivas (JIRA)" <ax...@ws.apache.org>.
[ http://nagoya.apache.org/jira/browse/AXIS-1608?page=history ]
Davanum Srinivas updated AXIS-1608:
-----------------------------------
Attachment: axisperftest.zip
Axis Performance test with bat files for both Axis 1.1 and Axis 1.2.
> Test case fro comparing Axis 1.2 against Axis 1.1
> -------------------------------------------------
>
> Key: AXIS-1608
> URL: http://nagoya.apache.org/jira/browse/AXIS-1608
> Project: Axis
> Type: Test
> Components: Basic Architecture
> Versions: 1.2RC2
> Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
> Reporter: Venkat Reddy
> Attachments: axisperftest.zip, axisperftest.zip, perftest.cs, perftest2.zip, perftest2.zip
>
> My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
> Timings (Average over 10 samples):
> ------------------------
> Axis 1.1 : 3891msecs
> Axis 1.2 : 5242 msec
> -- Venkat
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://nagoya.apache.org/jira/secure/Administrators.jspa
-
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Venkat Reddy
Created: Sun, 17 Oct 2004 10:09 PM
Body:
Dims,
Actually, for all my tests posted here, i have discarded initial four or five results, and taken only the last few results (which are close) for average. Please let me know if I still need to do the bootstrap call.
On the other note, using CommonsHTTPSender on client didn't improve the situation.
Thanks
Venkat
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54279
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Sun, 17 Oct 2004 10:09 PM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Venkat Reddy
Created: Sun, 17 Oct 2004 10:41 PM
Body:
Results with bootstrap call (Average over 10 samples).
Axis1.1: 3570 msecs
Axis1.2: 7208 msecs
I'll attach the updated test case.
-- Venkat
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54281
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Sun, 17 Oct 2004 10:41 PM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Updated: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following issue has been updated:
Updater: Dan Ciarniello (mailto:dciarniello@cityxpress.com)
Date: Thu, 14 Oct 2004 9:08 AM
Comment:
Since I'm the original reporter of this issue to the axis-users mailing list, I figured I'd add my test case. The attachment includes the WSDL, the service implemenation class and the test client. It's quite rough and just intended to give an idea of the performance difference.
The numbers I get are
Axis 1.1: ~2400 ms
Axis 1.2: ~5300 ms
I used WSDL2Java to generate both the server and client side classes.
Changes:
Attachment changed to axisperftest.zip
---------------------------------------------------------------------
For a full history of the issue, see:
http://issues.apache.org/jira/browse/AXIS-1608?page=history
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Thu, 14 Oct 2004 9:08 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
[jira] Commented: (AXIS-1608) Test case fro comparing Axis 1.2 against Axis 1.1
Posted by ax...@ws.apache.org.
The following comment has been added to this issue:
Author: Davanum Srinivas
Created: Thu, 14 Oct 2004 7:06 AM
Body:
AND please use CommonsHTTPSender in both Axis 1.2 and 1.1 clients. (as i does HTTP 1.1 - no need to compute the size of the message before sending it)
thanks,
dims
---------------------------------------------------------------------
View this comment:
http://issues.apache.org/jira/browse/AXIS-1608?page=comments#action_54044
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/AXIS-1608
Here is an overview of the issue:
---------------------------------------------------------------------
Key: AXIS-1608
Summary: Test case fro comparing Axis 1.2 against Axis 1.1
Type: Test
Status: Unassigned
Priority: Major
Project: Axis
Components:
Basic Architecture
Versions:
1.2RC2
Assignee:
Reporter: Venkat Reddy
Created: Thu, 14 Oct 2004 12:38 AM
Updated: Thu, 14 Oct 2004 7:06 AM
Environment: Win XP, 1 Gig Memory, 2.6 Ghz Intel, Axis 1.2 RC2, Axis 1.1 Final
Description:
My results show that Axis 1.2 RC (Latest CVS) is 1.35 times slower than Axis 1.1. The test is conducted using a simple web service client. Time for the round trip is measured and compared. Client sends a string of about 8 MB size to an RPC/encoded style web service deployed on Tomcat 5.0.25
Timings (Average over 10 samples):
------------------------
Axis 1.1 : 3891msecs
Axis 1.2 : 5242 msec
-- Venkat
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira