You are viewing a plain text version of this content. The canonical link for it is here.
Posted to c-dev@axis.apache.org by damitha kumarage <da...@opensource.lk> on 2004/10/12 11:54:40 UTC
Test with larger string arrays
Hi,
I added a test in tests/client/performance named large_arrays.
There I can send arbitrary larger strings in arbitrary large arrays,
arbitrary number of times. Attached is a result for testing with 256
strings of 1024 array size run thousand times. Attached also is the
readme file describing how to run the test
thanks
damitha
Re: Test with larger string arrays
Posted by damitha kumarage <da...@opensource.lk>.
Hi samisa,
Very sorry for the mistake I did. It seems that i do lot of mistakes
these days.
I intended to create seperated test inside performance folder. I started
that test initially copying from time folder into a new folder and
worked on it. But alas, I forgot to remove the CVS folder before I
commit my own stuff. As you can see this replace your work from my work,
because both folder contained same CVS folder locally. I corrected that
and now time and larger_strings are two seperated tests in two seperate
folders.
I really didn't intend to replace your work. I really really didn't mean
it.
Acturally the small java program is not related to the test directly. It
just create the data file. It's just a small utility.
thanks
damitha
On Tue, 2004-10-12 at 16:14, Samisa Abeysinghe wrote:
> Hi Damitha,
> I appreciate you thoughtfull work to help test arrays.
> However, with the CVS email, I notice that you have erased bulk part of my erlier code for
> time, including usage semantics (or am I mistaken here?). This is kind of disturbing. (as I have
> been saying, the usage semantics are quite important so that I could test both againt Axis C++ and
> Axis Java services)
> As these are small programs, I think it is better if you could add your stuff as a seperate
> client. Please see if you can make your code independent of mine and put the earlier client back
> inplace. (As far as I understood from the readme file, new code has different semantcs. I will try
> and see if I could sync up with that)
> Also, please let me know if you have issues if I put the service_url usage back in place.
>
> PS: Why have U used a java client here? I think we can do the same with a C++ client with ease :)
>
> Thanks,
> Samisa...
>
> --- damitha kumarage <da...@opensource.lk> wrote:
>
> > Hi,
> >
> > I added a test in tests/client/performance named large_arrays.
> >
> > There I can send arbitrary larger strings in arbitrary large arrays,
> > arbitrary number of times. Attached is a result for testing with 256
> > strings of 1024 array size run thousand times. Attached also is the
> > readme file describing how to run the test
> >
> > thanks
> > damitha
> > > time(micro secs taken each for 1000 runs of sending 1024 size string arrays
> > 128624
> > 180721
> > 181792
> > 178766
> > 175624
> > 176261
> > 174927
> > 177375
> > 177690
> > 173842
> > 173776
> > 172123
> > 169486
> > 168281
> > 170073
> > 167924
> > 167953
> > 170410
> > 171836
> > 168780
> > 172288
> > 170004
> > 168781
> > 169285
> > 168147
> > 171345
> > 169761
> > 167595
> > 167977
> > 171796
> > 168614
> > 169909
> > 170330
> > 170871
> > 167608
> > 169898
> > 172076
> > 170439
> > 172338
> > 167810
> > 170710
> > 168149
> > 167632
> > 168719
> > 168561
> > 167434
> > 168260
> > 171387
> > 174869
> > 181676
> > 181254
> > 180066
> > 176536
> > 171277
> > 940727
> > 352947
> > 306332
> > 320503
> > 128563
> > 138583
> > 130700
> > 126506
> > 126478
> > 127119
> > 126378
> > 126309
> > 127012
> > 128785
> > 127307
> > 125655
> > 126887
> > 125928
> > 127410
> > 126034
> > 126060
> > 126302
> > 127651
> > 126428
> > 126503
> > 126156
> > 126289
> > 126236
> > 126746
> > 126058
> > 127373
> > 126740
> > 125732
> > 126194
> > 389213
> > 131859
> > 979583
> > 323765
> > 173415
> > 282312
> > 271292
> > 170934
> > 213213
> > 168607
> > 171765
> > 169736
> > 170500
> > 168783
> > 172265
> > 172006
> > 168495
> > 172380
> > 169938
> > 171112
> > 170574
> > 169281
> > 169765
> > 169539
> > 170196
> > 172050
> > 170992
> > 171446
> > 170405
> > 170800
> > 174159
> > 168664
> > 172880
> > 169055
> > 170900
> > 172519
> > 169310
> > 171017
> > 168997
> > 169835
> > 168471
> > 168965
> > 169922
> > 171961
> > 169401
> > 171317
> > 170051
> > 170835
> > 169166
> > 170649
> > 169492
> > 170618
> > 169650
> > 168941
> > 172862
> > 167951
> > 168609
> > 172324
> > 168617
> > 173263
> > 168289
> > 171634
> > 170424
> > 171032
> > 168317
> > 172296
> > 171551
> > 170616
> > 170811
> > 168913
> > 168513
> > 168867
> > 171618
> > 170642
> > 168624
> > 170641
> > 170088
> > 172325
> > 170284
> > 170086
> > 171124
> > 169857
> > 169208
> > 170350
> > 171094
> > 170314
> > 169265
> > 171890
> > 168300
> > 172245
> > 172140
> > 170975
> > 168953
> > 168690
> > 172474
> > 168384
> > 169474
> > 170485
> > 170094
> > 170544
> > 168337
> > 172216
> > 171250
> > 170274
> > 169068
> > 171919
> > 171479
> > 171147
> > 168673
> > 170483
> > 169290
> > 171610
> >
> === message truncated ===> This test let you test with large string arrays.
> > String as well as the array size can be changed at run time
> > for testing. Also you can change the number of test runs.
> >
> > To create a data input file which contains strings of given maximum length
> > and given number of lines, execute the following. The array size corresponds
> > to String array size.
> >
> > format:java GetInput <maximum line length> <array size>
> >
> > eg:java GenInput 128 1024 > stringdata.txt
> >
> > If you want to test with larger strings, increase the maximum line length.
> > If you need to test with large array sizes, increase the array size.
> >
> > Now to run the sample using this generated stringdata.txt input file
> >
> > format:largeArrays <array size> <maximum line length> <number of runs> <datafile>
> > <comment>
> >
> > eg:./largeArrays 1024 128 100 result "time(micro secs) taken each for 100 runs of
> > sending 1024 size string arrays"
> >
> > array size should be less than or equal to number of lines in stringdata.txt.
> >
> > maximum line length should be greater than or equal to the maximum string size
> > in stringdata.txt.
> >
> > number of runs is the number of times you need to run the test
> >
> > datafile is the stringdata.txt you created in the first step.
> >
> > comment you put will appear in the head of the result file generated at the
> > end of the test run.
> >
>
>
>
>
> _______________________________
> Do you Yahoo!?
> Declare Yourself - Register online to vote today!
> http://vote.yahoo.com
>
Re: IPv6
Posted by Samisa Abeysinghe <sa...@yahoo.com>.
Hi Paul,
No I vae not looked into IPV6.
Hopefully this could be made a 1.4 feature if required.
Thanks,
Samisa...
--- Paul Fremantle <pz...@hursley.ibm.com> wrote:
> Samisa
>
> Have you looked at IPv6 for your transport? (Its just a question not a
> hint!)
>
> Paul
>
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com
IPv6
Posted by Paul Fremantle <pz...@hursley.ibm.com>.
Samisa
Have you looked at IPv6 for your transport? (Its just a question not a
hint!)
Paul
Re: Test with larger string arrays
Posted by Samisa Abeysinghe <sa...@yahoo.com>.
Hi Damitha,
I appreciate you thoughtfull work to help test arrays.
However, with the CVS email, I notice that you have erased bulk part of my erlier code for
time, including usage semantics (or am I mistaken here?). This is kind of disturbing. (as I have
been saying, the usage semantics are quite important so that I could test both againt Axis C++ and
Axis Java services)
As these are small programs, I think it is better if you could add your stuff as a seperate
client. Please see if you can make your code independent of mine and put the earlier client back
inplace. (As far as I understood from the readme file, new code has different semantcs. I will try
and see if I could sync up with that)
Also, please let me know if you have issues if I put the service_url usage back in place.
PS: Why have U used a java client here? I think we can do the same with a C++ client with ease :)
Thanks,
Samisa...
--- damitha kumarage <da...@opensource.lk> wrote:
> Hi,
>
> I added a test in tests/client/performance named large_arrays.
>
> There I can send arbitrary larger strings in arbitrary large arrays,
> arbitrary number of times. Attached is a result for testing with 256
> strings of 1024 array size run thousand times. Attached also is the
> readme file describing how to run the test
>
> thanks
> damitha
> > time(micro secs taken each for 1000 runs of sending 1024 size string arrays
> 128624
> 180721
> 181792
> 178766
> 175624
> 176261
> 174927
> 177375
> 177690
> 173842
> 173776
> 172123
> 169486
> 168281
> 170073
> 167924
> 167953
> 170410
> 171836
> 168780
> 172288
> 170004
> 168781
> 169285
> 168147
> 171345
> 169761
> 167595
> 167977
> 171796
> 168614
> 169909
> 170330
> 170871
> 167608
> 169898
> 172076
> 170439
> 172338
> 167810
> 170710
> 168149
> 167632
> 168719
> 168561
> 167434
> 168260
> 171387
> 174869
> 181676
> 181254
> 180066
> 176536
> 171277
> 940727
> 352947
> 306332
> 320503
> 128563
> 138583
> 130700
> 126506
> 126478
> 127119
> 126378
> 126309
> 127012
> 128785
> 127307
> 125655
> 126887
> 125928
> 127410
> 126034
> 126060
> 126302
> 127651
> 126428
> 126503
> 126156
> 126289
> 126236
> 126746
> 126058
> 127373
> 126740
> 125732
> 126194
> 389213
> 131859
> 979583
> 323765
> 173415
> 282312
> 271292
> 170934
> 213213
> 168607
> 171765
> 169736
> 170500
> 168783
> 172265
> 172006
> 168495
> 172380
> 169938
> 171112
> 170574
> 169281
> 169765
> 169539
> 170196
> 172050
> 170992
> 171446
> 170405
> 170800
> 174159
> 168664
> 172880
> 169055
> 170900
> 172519
> 169310
> 171017
> 168997
> 169835
> 168471
> 168965
> 169922
> 171961
> 169401
> 171317
> 170051
> 170835
> 169166
> 170649
> 169492
> 170618
> 169650
> 168941
> 172862
> 167951
> 168609
> 172324
> 168617
> 173263
> 168289
> 171634
> 170424
> 171032
> 168317
> 172296
> 171551
> 170616
> 170811
> 168913
> 168513
> 168867
> 171618
> 170642
> 168624
> 170641
> 170088
> 172325
> 170284
> 170086
> 171124
> 169857
> 169208
> 170350
> 171094
> 170314
> 169265
> 171890
> 168300
> 172245
> 172140
> 170975
> 168953
> 168690
> 172474
> 168384
> 169474
> 170485
> 170094
> 170544
> 168337
> 172216
> 171250
> 170274
> 169068
> 171919
> 171479
> 171147
> 168673
> 170483
> 169290
> 171610
>
=== message truncated ===> This test let you test with large string arrays.
> String as well as the array size can be changed at run time
> for testing. Also you can change the number of test runs.
>
> To create a data input file which contains strings of given maximum length
> and given number of lines, execute the following. The array size corresponds
> to String array size.
>
> format:java GetInput <maximum line length> <array size>
>
> eg:java GenInput 128 1024 > stringdata.txt
>
> If you want to test with larger strings, increase the maximum line length.
> If you need to test with large array sizes, increase the array size.
>
> Now to run the sample using this generated stringdata.txt input file
>
> format:largeArrays <array size> <maximum line length> <number of runs> <datafile>
> <comment>
>
> eg:./largeArrays 1024 128 100 result "time(micro secs) taken each for 100 runs of
> sending 1024 size string arrays"
>
> array size should be less than or equal to number of lines in stringdata.txt.
>
> maximum line length should be greater than or equal to the maximum string size
> in stringdata.txt.
>
> number of runs is the number of times you need to run the test
>
> datafile is the stringdata.txt you created in the first step.
>
> comment you put will appear in the head of the result file generated at the
> end of the test run.
>
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com