of gathering data: sampling and
instrumenting:
sampling is less intrusive but will take longer to get good covereage,
whereas instrumenting has limits on the methods it can profile and makes
a huge impact on overall performance,
(which therefore invalidates the results a little bit).
Most
() and serialize() methods directly to the
classes generated by wsdl2java. So mostly pure performance of bean and
array serializers and deserializers have been measured in the benchmark.
Multithreading has been implemented using a thread that is doing
repeatedly serialization and deserialization
Using the latest source code (ant build from today's source download)
(will be called New code) and the Axis 1.2 RC2 (will be called Old
code), I run the test with different parameters. The client side is 1
application with 1 or more threads. Each thread makes a number of
requests to the
[mailto:[EMAIL PROTECTED]
Sent: 10 January 2005 17:39
To: [EMAIL PROTECTED]
Subject: Re: axis performance in multithreading env
Using the latest source code (ant build from today's source download)
(will be called New code) and the Axis 1.2 RC2 (will be called Old
code), I run the test
This is a problem for me. I use Netbeans and the JFluid supports 1.4,
not Jdk1.5. My apps run under 1.5. It would take some effort on my
part to make these apps run elsewhere. I am currently tied up at a
moment, so I have to put this off for now. I am sorry about this. If
time permit,
-patched version. Performance goes slight down with the patch, but
not significant either.
Again, could you please tell me how was your set up of your test.
For sanity check, I download the binary at:
http://cvs.apache.org/dist/axis/nightly/axis-bin-1_2RC2.zip
Notice that the binary is stilled label
Thanks. I can't wait to get my hand on this. Well, I'll get the
nightly build and try it out.
What's the status of this? This would be really nice if it's
incorporated into the next release candidate of Axis.
On Mon, 06 Dec 2004 16:40:34 +0100, [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
Hi,
While doing performance tests on deserialization, I noticed that Axis
scales very poorly when
I tried to send corrupted message to an Axis server. Normally, I would
get ~ 250 hits/seconds. However, when a corrupted message is sent to
the server, I get 1 hit per 9 seconds. This is more than 2 thousand
times slowser!
Could this be used for denial of service attack? Also, because the
to
[EMAIL PROTECTED]
To
[EMAIL PROTECTED], John Walker [EMAIL PROTECTED]
cc
Subject
Re: Attachment Performance
James,
your disk might have filled up ! clear the /tmp/
folder where the attachments are stored written to.
Mayur
--- John Walker [EMAIL PROTECTED] wrote:
James,
Which version
:
Hi.
Recently,when using Axis, I've found that the
performance of the
getAttachments method has disimproved.
It's been fine up to now, and for no real reason,
getting the Iteration of
attachments while reading
Hi.
Recently,when using Axis, I've found that the performance of the
getAttachments method has disimproved.
It's been fine up to now, and for no real reason, getting the Iteration of
attachments while reading a message with one atttachment (of about
600Bytes) takes 30 seconds.
Does
] wrote:
Hi.
Recently,when using Axis, I've found that the
performance of the
getAttachments method has disimproved.
It's been fine up to now, and for no real reason,
getting the Iteration of
attachments while reading a message with one
atttachment (of about
600Bytes) takes 30 seconds
Hi,
While doing performance tests on deserialization, I noticed that Axis
scales very poorly when increasing the number of threads.
Here are the figures:
5 threads, 1000 messages: 15 seconds
10 threads, 1000 messages: 56 seconds
15 threads, 1000 messages: 93 seconds
The root of the problem
Could you please open a bug report with a diff -u against latest
XMLUtils.java?
thanks,
dims
On Mon, 06 Dec 2004 16:40:34 +0100, [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
Hi,
While doing performance tests on deserialization, I noticed that Axis
scales very poorly when increasing
parameter first, and then into attachments. The idea is that if the file
is small (ie: 100 KB) then using the parameter is ok, otherwise the
attachment got to be used.
The problem is that we observe unexpected performance in production. We
have built some jmeter load tests, trying to send 4 MB
After some more thoughts it looks that we are not doing buffered I/O
while reading the attachment from disk. As suggested in many places (for
example
http://java.sun.com/docs/books/performance/1st_edition/html/JPIOPerformance.fm.html)
I/O operations in java are by default unbuffered.
Shame
I was looking for a way to increase performance, because it seems the
server is topped off at around 100 request per seconds. Searching for
information about keep alive seem not be so easy. But at the end, I get
it to work by doing this:
1) Download the commons-httpclient at apache.org
2
Thank you for your suggestion. The client and server
are not on the same machine. I can't modify the server
side, just query it with soap or axis. I tried 3 ways
to connect :
- IP address in URL constructor : 15230 ms
- logical name, using /etc/hosts : 15180 ms
- bad logical name : 13256 ms to
Bernard,
Axis is like any other web application. Depending on how
it is deployed, the first service call may have to wait for the application
server to create an instance of the Axis servlet and initialize it. In
addition, I guess Axis is doing something similar for each service that
is
Thank you Tony for your explanation, but my problem is
only on the client side, since the measurements are
good when I'm using an Apache Soap 2.3.1 client.
The total time for my calls are :
ClientServerFirst callSubsequent calls
Soap Soap 3653 ms 958 ms
Axis
How about try to add the server's ip/address to the client machine's
host file. Event if it's the same machine, add localhost in anyway.
Please let me know if that solves the initial delay.
Bernard LUPIN wrote:
Hi all,
I'm writing a client that will query web services
already developped with
Hi,
In term of performance and scalability, if my business logic is a java
class and I want to expose it as a web service.
Is it better to add an ejb and expose it as a web service (This new ejb
merges the business API and forwards calls to the java class. It uses
services provided by the ejb
]
Subject: Web Service performance
Hi,
In term of performance and scalability, if my business logic is a java
class and I want to expose it as a web service.
Is it better to add an ejb and expose it as a web service (This new ejb
merges the business API and forwards calls to the java class
Hi all,
I'm writing a client that will query web services
already developped with Apache Soap 2.3.1.
I tested two versions of my client, one with Apache
Soap 2.3.1 and one with Axis 1.2RC1.
The return values are good with both clients, but the
execution time is 3 seconds with Soap 2.3.1, and 15
, Davanum Srinivas [EMAIL PROTECTED] wrote:
Please log a jira bug witjh your test case and performance numbers.
thanks,
dims
On Wed, 13 Oct 2004 13:16:01 -0700, Dan Ciarniello
[EMAIL PROTECTED] wrote:
I've taken a very simple web service and deployed it using Axis 1.1 and
Axis 1.2
Perhaps this paper can give you guys some help on evaluating Web
Services implementations.
http://groups-beta.google.com/group/qworks/browse_thread/thread/81842355c8f0472e
--
Qworks
http://groups-beta.google.com/group/qworks
I've taken a very simple web service and deployed it using Axis 1.1 and
Axis 1.2 (RC1) on Tomcat 4.1.30 on a Linux machine (Fedora Core 1) and
done some rough performance tests. My tests show that Axis 1.2 takes at
least twice as long to process requests as Axis 1.1. This is of great
concern
Please log a jira bug witjh your test case and performance numbers.
thanks,
dims
On Wed, 13 Oct 2004 13:16:01 -0700, Dan Ciarniello
[EMAIL PROTECTED] wrote:
I've taken a very simple web service and deployed it using Axis 1.1 and
Axis 1.2 (RC1) on Tomcat 4.1.30 on a Linux machine (Fedora Core
Title: Performance issue with axis deserialization
Hi,
I have hosted a webservice using axis on tomcat 4.1.24.
I m finding that as concurent load increases, there is a tremendous growth in the deserialization time taken.
Any clues ??
Thanks
aravind
Hi all axis user,
I would like to know what happens when we call the first time a service.
To be easily understood, here is an example :
I've got one client C1.
One service WS1 deployed on the server S1.
I start running my server :
S1 - Application axis (AXIS) initialized...
Then, C1 call the
The Axis servlet loads web services on request,
depending on the scope. As you have session scope, the web service class
will be instantiated once for each session. So the first call, in that
session, will take longer than subsequent calls.
I'm not aware of any configuration that will
cause Axis
Hi,
I'm implementing my first Web Service and I have a question regarding the
scope of the Web Service and its influence on the performance: my Web
Service won't have any state (local variables), will only perform read
operations (therefore I don't have any locking problem) and is going to be
use
there is no
requirement to manage state across method calls for a particular
user). I think it is better to go for application scope since u have
performance in mind as well. If u have request scope, then a new
instance of the web service object will be created to serve each
request (each method call
any better
alternatives. Keep us posted if you find more!
Keith
-Original Message-
From: Priest, Mark [mailto:[EMAIL PROTECTED]
Sent: 23 August 2004 15:53
To: '[EMAIL PROTECTED]'
Subject: Class/resource loading with applet killing performance
Hello,
We are having an issue
a GET
request to the web server, which burns up a lot of bandwidth. There are
three types of things that the Axis runtime is trying to load that are
causing us performance problems:
1. _Helper classes. We don't use any _Helper classes but Axis tries to find
them anyway
2. ResourceBundle Strings
The performance improved when we copied all the jars the client depends on
to one folder and executed the client in the following way. This again is
not consistent across two linux machines. But if anybody has the idea,
please do let me know.
java -Djava.endorsed.dirs
/2004 08:00 cc
PM
Subject
Re: Performance issue with using
Please
PROTECTED]
Sent: Friday, August 06, 2004 10:13 AM
To: [EMAIL PROTECTED]
Subject: Re: Performance issue with using Axis
It's class loader if the creation of a new locator took that long. I
strongly think that is not the case. Chance is very high that your
networking configuration is bad. Check
Subject
Re: Performance issue with using
Please respond to Axis
[EMAIL PROTECTED]
he.org
Re: Performance issue with using
Please respond to Axis
[EMAIL PROTECTED]
he.org
Re: Performance issue with using
Please respond to Axis
[EMAIL PROTECTED]
he.org
Check your network connection settings, such as hosts file, etc. Other
than that, you can find a hello world tutorial for axis somewhere
complete with client and server to make sure there's nothing wrong with
the way you do it. 5 seconds long is just plain wrong, so definitely
you're having
Hi Hrishikesh: I would be happy to offer advice on how to improve the
system's performance. I recommend you take a look at the
http://www.pushtotest.com site for a list of resources and articles on
improving SOA scalability and performance. Additionally, if you would
send me the WSDL I would
Subject
Re: Performance issue with using
Please respond to Axis
[EMAIL PROTECTED
cc
PM
Subject
Re: Performance issue with using
Please respond to Axis
can
improve the performance.
Any tips will be appreciated and I can answer any questions if you have.
Thanks,
Hrishikesh.
on the client side we are thinking that
it is causing the problem. Can you please guide me where and how we can
improve the performance.
Any tips will be appreciated and I can answer any questions if you have.
Thanks,
Hrishikesh.
PROTECTED]
08/03/2004 10:22 cc
AM
Subject
Re: Performance
Hi;
Does anyone know of any performance numbers out there for
performing xpath and/or xslt to pull nodes out of xml where the xml is on the
order of 100 megabytes of data?
Thanks - dave
*excellent* This link is going in my keeper list. Naturally now you'll
have to keep this list up to date, right? :)
Jim Murphy
Mindreef, Inc.
Dennis Sosnoski wrote:
I've finally published my performance comparison of SOAP web services
using open source frameworks, at
http
this by default, as
does the current JibxSoap). I can also try turning off sendXsiTypes for
the rpc/enc version, which should help its performance. Anyone know of
any other flags or controls that might make a significant difference for
Axis?
- Dennis
Jim Murphy wrote:
*excellent* This link
I've finally published my performance comparison of SOAP web services
using open source frameworks, at
http://www.sosnoski.com/presents/cleansoap/comparing.html In addition to
the performance comparison itself (using JAX-RI doc/lit, Axis doc/lit,
Axis-Castor, Axis rpc/enc, JibxSoap... and RMI
Dennis,
We have not yet tuned performance for this release :( Was planning to
do it after the beta. Yes, i'd like to use the examples to help tune
Axis.
thanks,
dims
On Fri, 09 Jul 2004 15:22:49 -0700, Dennis Sosnoski [EMAIL PROTECTED] wrote:
I've finally published my performance comparison
] wrote:
I've finally published my performance comparison of SOAP web services
using open source frameworks, at
http://www.sosnoski.com/presents/cleansoap/comparing.html In addition to
the performance comparison itself (using JAX-RI doc/lit, Axis doc/lit,
Axis-Castor, Axis rpc/enc, JibxSoap
not sure
about - it may be that it'll run all three, except for needing to change
a name or two (but it doesn't include the Castor serializers/deserializers).
I'd seen some traffic from you on the list discussing some performance
improvements last month, so that was why I thought you might have made
FYI, Performance Index of public web services - http://wspi.ca.com/
--
Davanum Srinivas - http://webservices.apache.org/~dims/
Folks,
If you are facing a performance problem(s), please try latest CVS /
nightly, if the problem still exists, open up a bug report with a
stripped down sample to recreate the problem. Am planning to get back
to some performance testing next week. If we can't recreate it, we
can't fix
Hello everyone,
I am starting performance testing on my Axis web service and I wanted to know if there
are any
java based performance test tools that measure concurrent clients. Any ideas?
Hut
The ones I can think of are Load Runner and Jprobe. Can't remember who
the vendors are.
-Original Message-
From: Hut Carspecken [mailto:[EMAIL PROTECTED]
Sent: Thursday, May 27, 2004 10:55 AM
To: [EMAIL PROTECTED]
Subject: Open Source Performance Test Tool That Measures Concurrent
Performance Test Tool That Measures Concurrent
Clients
Hello everyone,
I am starting performance testing on my Axis web service and I wanted to
know if there are any java based performance test tools that measure
concurrent clients. Any ideas?
Hut
You can probably use JMeter to emulate concurrent clients.
Phil
-Original Message-
From: Miller, Janet [mailto:[EMAIL PROTECTED]
Sent: 27 May 2004 16:04
To: [EMAIL PROTECTED]
Subject: RE: Open Source Performance Test Tool That Measures
Concurrent
Clients
Woops. Sorry. After
Hi folks,
I need help testing latest cvs. Added a performance/volume test case that sends
100,000 strings
(original problem description - http://issues.apache.org/jira/browse/AXIS-1323). If
you wish to
help:
#1: get latest cvs
#2: run ant clean in ws-axis/java directory
#3: run ant in ws-axis
On Apr 26, 2004, at 7:36 AM, Davanum Srinivas wrote:
Hi folks,
I need help testing latest cvs. Added a performance/volume test case
that sends 100,000 strings
(original problem description -
http://issues.apache.org/jira/browse/AXIS-1323). If you wish to
help:
#1: get latest cvs
#2: run ant
plz remove xmlsec.jar from your env and try again.
--- Joe Nall [EMAIL PROTECTED] wrote:
On Apr 26, 2004, at 7:36 AM, Davanum Srinivas wrote:
Hi folks,
I need help testing latest cvs. Added a performance/volume test case
that sends 100,000 strings
(original problem description
Folks,
Off and on we get queries about performance problems. So here's a request. If you have
ANY
concrete examples of performance problems, PLEASE open a bug report ASAP with enough
sample code
to recreate the problem. If we can't recreate it, we can't fix it. period.
Thanks,
dims
Hello everyone...
as you are just talking about performance issues, I have one ;o)
I would be happy to get some solution to the problem I explaned here:
http://marc.theaimsgroup.com/?t=10790179657r=1w=2
I will provide more information if anyone needs it.
Greetings
- Felix
plz log a bug with your sample code (Make sure you are using Axis 1.2 Beta)
-- dims
--- Felix Dierich [EMAIL PROTECTED] wrote:
Hello everyone...
as you are just talking about performance issues, I have one ;o)
I would be happy to get some solution to the problem I explaned here:
http
Hi all,
Has anybody done performance improvements with SOAP
calls?
IN addition, how would I make sure that a
call.invoke opens and
CLOSES the connection. I want to make sure that the
connection is
not open.
Does the call.setTimeout work?
Any input in this regard is
appreciated
: Thursday, April 15, 2004 9:29 PM
To: [EMAIL PROTECTED]
Subject: Performance problem
(production) urgent
Hi all,
Has anybody done performance
improvements with SOAP calls?
IN addition, how would I make sure
that a call.invoke opens and
CLOSES the connection. I want to
make
Hi,
Has anyone seen any comparison documents between different web services
toolkits? I have been training the use of AXIS for few weeks now and I
have liked it. But today I heard that I should stay away from AXIS
because it has some performance issues. The person who told that, didn't
want
because it has some performance issues. The person who told that, didn't
want to explain those issues.
I would like to find comparisons between toolkits and some performance
statistics.
So if anyone could help me, it would be really nice.
Jani
=
Davanum Srinivas - http
] wrote:
Hi,
Has anyone seen any comparison documents between different web services
toolkits? I have been training the use of AXIS for few weeks now and I
have liked it. But today I heard that I should stay away from AXIS
because it has some performance issues. The person who told that, didn't
again my question:
Which is the better way for developing Web Services to use with existing
JavaWebStartClient and EJBs.
Has the result of both ways the same performance?
Thorsten
(methodName);
Object ret = call.invoke(new Object[]{});
Now again my question:
Which is the better way for developing Web Services to use with existing
JavaWebStartClient and EJBs.
Has the result of both ways the same performance?
Thorsten
~~
Anne Thomas Manes
VP Research Director
Hi,
i'm relatively new to Axis and have a working connection from JavaWebStart
Client to a SessionBean on my JBoss Appliation Server with WSDD.
Which advantages has using WSDL instead of WSDD. For me as newby it seems
to be much more complicated to create a running System. Is the performance
JavaWebStart
Client to a SessionBean on my JBoss Appliation Server with WSDD.
Which advantages has using WSDL instead of WSDD. For me as newby it seems
to be much more complicated to create a running System. Is the performance
better or the administration of the webservices or what else?
Thanks
Jim Murphy wrote:
2.2.1: Not sure what you mean by start and stop. Not knowing anything
else I'd put my money on the perf bottleneck being at inter-app
communication. In a servlet container (esp. Tomcat) things can run
inprocess. So there is not extra parameter marshaling that has to
Hi Vivek!
Oracle's JDeveloper has very good and useful profiling options. You can get it for
free (only for development purposes - what else would you do with it?) at
otn.oracle.com. Takes a little fiddling into, but is very powerful. It can profile
down into any level, as long as you include
Hello All,
I am looking for tools/profilers to track the performance of a web service
in Axis. Specifically, I would be interested in collecting statistics for
memory processing, time taken to serve a request, to name few..
If I can get the time taken to (De)serialize a SOAP request would
I expect that https will give you better performance than ws-security,
but even if you're only using http, the functionality is not the same.
https gives you point-to-point message security, but it does not
effectively support end-to-end security, e.g., strong authentication,
authorization
Hi all!
Thanks for your suggestion Jay!
Caching the Service object by keeping a reference would definitely help,
I think, but as my clients are called by command line and always start a
new VM for architecture reasons, I can't keep the Service object in
memory. I need a way to make the first
is users!I have developed a rather large project with Apache Axis and liked Axisvery much so far (with the occasional problem of missing documentation).I have a performance problem with my client applications though: Thefirst call to the web service takes almost two seconds on rather fastmachines. The
Hello Axis users!
I have developed a rather large project with Apache Axis and liked Axis
very much so far (with the occasional problem of missing documentation).
I have a performance problem with my client applications though: The
first call to the web service takes almost two seconds on rather
msec
Average time for invocation of the same service for 5 clients for 10 times
---~600msec
Could you please tell me as to where it went wrong and tips to improve the
performance? Are there any document regardin this?
Thanks in advance.
Thanks regards,
Ayyappan Gandhirajan
Hi list,
I'm playing with soap attachments in dime format using axis.
I want to measure performance between soap implementation and custom
implementation i.e
sending binary fail over http. In my test case soap is about twise
slower then plain http.
Could someone point mi a way of increasing
Hello,
We are using Apache SOAP as our current WebService
engine. Since the workload of the server
has dramaticly increased we are planing to migrate
to Apache Axis in hope for better performance.
Are there any profe of this?
/Lars
How about unplugging your dev workstation from the Net and see if the
response time differs?
/Chris
http://cvs.apache.org/~haddadc
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent: Monday, October 27, 2003 7:54 PM
To: [EMAIL PROTECTED]
Subject: axis performance
:[EMAIL PROTECTED]
Sent: Monday, October 27, 2003 7:54 PM
To: [EMAIL PROTECTED]
Subject: axis performance question
Hi,
I have got web service clients (both Java and VB) that need to access my
Axis Web Service.
The web service is located in my intranet.
Running the client from my workstation, I get
Hi,
I have got web service clients (both Java and VB) that need to access my
Axis Web Service.
The web service is located in my intranet.
Running the client from my workstation, I get relatively good performance.
Once I move the client to our UAT box which has no internet access
Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent: Friday, October 03, 2003 2:09 PM
To: [EMAIL PROTECTED]
Subject: Performance Issues with AXIS
Importance: High
Hi,
We are using Apache AXIS 1.1 to support a web service that returns images.
The images are base-64 encoded and embedded
The Xerces code is very similar. What's the effect if you take the
synchronize out and rerun your performance tests? Do you have any
numbers or graphs?
-Original Message-
From: Sutton, Ray [mailto:[EMAIL PROTECTED]
Sent: Thursday, October 09, 2003 1:32 PM
To: '[EMAIL PROTECTED
My personal instinct, unless anyone knows a good reason not, would be to
remove this specific synchronization point.
The docs are quite specific that a DocumentBuilderFactory is *not*
guaranteed to be thread-safe:
http://java.sun.com/j2se/1.4.2/docs/api/javax/xml/parsers/DocumentBuilderFac
PROTECTED]
Sent: Thursday, October 09, 2003 12:59 PM
To: [EMAIL PROTECTED]
Subject: RE: Performance Issues with AXIS Axis Response Time appears
linear with load
The Xerces code is very similar. What's the effect if you take the
synchronize out and rerun your performance tests? Do you have any
: Sutton, Ray [mailto:[EMAIL PROTECTED]
Sent: Thursday, October 09, 2003 2:15 PM
To: '[EMAIL PROTECTED]'
Subject: RE: Performance Issues with AXIS Axis Response Time appears
linear with load
In answer to Mike, no haven't got that far yet, in light of Robert's
reply I will probably look for another
Hi,
I am getting performance problems of axis 1.1. It is running under Tomcat 4.1.2. After
the system has run for a couple of minutes during which the performance is excellent
the systems performance falls and at one point it is very slow. When i try to access
the wsdl file it also takes
What about if you try to send the images with SOAP attachments?
Patrick.
[EMAIL PROTECTED] wrote:
Hi,
We are using Apache AXIS 1.1 to support a web service that returns
images. The images are base-64 encoded and embedded within the XML.
What we are seeing is very slow response times with
I have integrated Tomcat 4.1.12 with Apache Axis 1.1. A few minutes after logging into
UI the server becomes very slow. It becomes so slow that even when i type the path to
the wsdl file in a browser it takes 25-30 seconds to bring up the wsdl file. Any
suggestion would be very appreciated.
What is your log4j root setting at? If it is at DEBUG, this could cause
issues.
On Fri, 2003-10-03 at 15:14, Abhijat Thakur wrote:
I have integrated Tomcat 4.1.12 with Apache Axis 1.1. A few minutes after logging
into UI the server becomes very slow. It becomes so slow that even when i type
there might be version problem.
-Original Message-
From: Benjamin Tomasini [mailto:[EMAIL PROTECTED]
Sent: Friday, October 03, 2003 12:39 PM
To: [EMAIL PROTECTED]
Subject: Re: performance problems
What is your log4j root setting at? If it is at DEBUG, this could cause
issues.
On Fri
1 - 100 of 202 matches
Mail list logo
