Hi,I'm currently developing a set of baseline tests which aims to measure performance changes different versions of software on our test systems. At the moment we have a skeleton set of tests which…Continue
A good approach is to run the requests you are trying to model through fiddler.
Then run JMeter through fiddler and compare the difference.
My tips are use a cookie manager, and a header manager.
Check the values you are passing…"
"I would say a useful skill is to understand pattern matching using regular expressions. Understanding of regular expressions will allow you to be more flexible when using other tools that facilitate these, JMeter, Wireshark etc... and would…"
":D - sorry about that. - we have 'automation monkeys' and 'code monkeys' as well. :-)
I guess it depends on the way you develop software. I have worked with companies in the past where they do review the spec…"
In my experience, I would say that automation can be extremely useful for developing regression packs that can run as part of a continuous integration cycle. They are useful for providing a snapshot and a confidence level for how a…"
You are more than welcome. I know what it is like when starting out. Fortunately I had some good mentor's to work with, and made some industry friends like Scott who have helped me out a lot over the years. Just…"
By data mining, I mean I would need to inspect the production database and extract information inside the database and try and build an understanding of what configuration and usage profile of a customer.
Thanks for your feedback…"
I will try to share some ideas, although I’m not sure how relevant they are as you don’t provide enough details of your system under test. For example I’m totally confused as to how you could “data mine…"
"When trying to come up with a model for performance tests, especially of an already live sysytem, I have generally tried to make a mix of your three bullets, with one slight difference.
I tend to try to talk to Customer Support (or the…"
"My approach to this problem would usually be to talk to various stakeholders and identify a set of business critical [load generating] processes. These can then be modelled and used to define a series of performance test cycles.
How you proceed from…"
Hi,I'm currently developing a set of baseline tests which aims to measure performance changes different versions of software on our test systems. At the moment we have a skeleton set of tests which test a small subset of the system. These are tests are not an accurate representation of live environment. The challenge is to improve these tests and try to model a subset of the live system.I’m wondering on which approach I should use when creating a set of models?My thoughts so far:1. Look at a…See More
Just to add, I do take your points on board David and apologise. I didn't deliberately intend to run a bypass of your question and responses, I stumbled on the link whilst reading blogs and genuinely thought the set of questions was…"
All registrations go through moderation. To improve quality of our members and avoid dreaded spammers.
Interested in software engineering, design patterns, algorithms, software testing, futurism / buddhism, open source / linux - and reading up on technology, Reading books on tech -> Bruce Snieder, James Whittaker, Kevin Mitnick, Bill Gates, Steven Levy, Richard Stallman.
-Try to keep the mind fresh and adapt to change and new ideas.