please, advise how can i achieve the following: You should eliminate "Think Time" from graph/stats You should eliminate "Pacing" (if tool counts) from graph/stats hall we need to eliminate ramp up and ramp down time for getting the average response time?
1. There is an option to eliminate "Think Time" duration in every analysis tool. If you specify the tool name then I would show you where this option is available.
2. Generally, pacing does not considered by LoadRunner and JMeter, so you need not to worry if you are conducting test with these tool. If you are using any other tool please let me know I will check and let you know.
3. Yes, only steady state duration (when all the users are ramp-up) should be considered for analysis; unless there is no error during ramp-up period.
Hi Gagan, In regard to think time I am having a query like I have defined think time in global.h (TT=15;) and add between two consecutive transactions. In scenario I have given the option like it should vary between 100% to 150%.It it correct or it will have a high impact on the result analysis.(load Runner 12.56)
Thanks for your post you will define every thing in easy way.
ReplyDeleteThanks Vishal. Please refer Level-II and Level III analysis for more understanding.
DeleteNice information and easy understanding
ReplyDeleteHi Gagan,
ReplyDeleteShall we need to eliminate ramp up and ramp down time for getting the average response time?
Yes, it is a good practice to analyse the performance metrics in steady state only.
DeleteThanks a lot for the article and the whole blog.
ReplyDeleteplease, advise how can i achieve the following:
You should eliminate "Think Time" from graph/stats
You should eliminate "Pacing" (if tool counts) from graph/stats
hall we need to eliminate ramp up and ramp down time for getting the average response time?
Hi Maha,
Delete1. There is an option to eliminate "Think Time" duration in every analysis tool. If you specify the tool name then I would show you where this option is available.
2. Generally, pacing does not considered by LoadRunner and JMeter, so you need not to worry if you are conducting test with these tool. If you are using any other tool please let me know I will check and let you know.
3. Yes, only steady state duration (when all the users are ramp-up) should be considered for analysis; unless there is no error during ramp-up period.
Thank you Gagan for the article,
DeleteCan you pls share any article for eliminating the 'Think Time' duration from JMeter tool.
Hi Avinash,
DeleteJMeter does not consider Think Time (Timer Duration).
Hi Gagan,
ReplyDeleteIn regard to think time I am having a query like I have defined think time in global.h (TT=15;) and add between two consecutive transactions. In scenario I have given the option like it should vary between 100% to 150%.It it correct or it will have a high impact on the result analysis.(load Runner 12.56)
Hi,
ReplyDeleteFirst thing,it is a good practice to keep random think time within a range.
Secondly, you can modify the range from 50% to 150%, so that the chances of breaching the TPS will be less.
In addition, I would suggest to use dynamic pacing option which will adjust the extra think time and will help to achieve the required TPS.