Measuring "how long" a particular feature takes to execute from front end perspe

Home Forums Software Testing Discussions Measuring "how long" a particular feature takes to execute from front end perspe

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #5935
    Erik
    Participant
    @ehendin

    Are there any tools out there that you folks would recommend for measuring “how long” a particular feature in a web application takes to execute from a front end perspective?

    I have seen some tools that claim to do this but do not seem to actually perform the action in the browser, or skip testing javascript that is executed in the process. So one tool I have used where the test will involve writing data but the data is not actually written when I replay the script. Yet it still “claims” to be showing performance statistics. Is that standard?

    We currently use a selenium framework with java webdriver/testNG and this has been great for functional verification(I spent years learning how it all works together). However often there are time delays in the script that would skew results.

    Any front-end tool that you guys would recommend? If not, then has anyone used Jmeter, and how difficult is it to set up and run – I downloaded it but it looks like I have to install it as an eclipse project.

    Any suggestions are most appreciated. My focus really is on performance with a single user, although at some point when my group’s environment is more set up for it, load will become much more of a consideration.

    Ideally this would be something that a non-programmer could use that does not cost a fortune because I am not sure what budget my company has.

    Thanks for reading.

    #5938
    Padmaraj
    Participant
    @padmaraj

    @Erik ( ehendin )

    You can follow my steps to get results:

    Load testing principles
    Step1: User scenario recording (request & response is recoded) & verified.
    Step2: Now specify the number of virtual user, from location & (user) browser emulate.
    Step3: Specify test duration – how much time you like to test.
    Step4: Simulate think time – how many minutes you spend on page.
    Step5: Specify Quality of Service – define acceptance criteria (page load time, time to first byte metrics)
    Step6: Results : Key matrices.
    Response time, throutput, performance of individual objects on page, slow pages on website.

    Analysis the report(reading & understanding report you need lot of technical information) & forward reports to developer / server infrastructure manager with your comments on what they need to be improved & wait for feedback and Re test again with new changes.

    Recommendation of tools from my side is Apache JMeter & LoadComplete.

    #5967
    Erik
    Participant
    @ehendin

    Thanks, Padmaraj!
    I have used Neoload and am quite familiar with the principles – but I found that steps executed in test recorded were not executed by the script and the support was not very helpful for the product.

    Again with JMeter I am not sure how much coding background is needed, but I will certainly look into LoadComplete. This is something I want my team to be able to use who may not have all have a coding background like myself.

    I also want to look into Mac-friendly options – I am noticing most of the solutions seem to be for pc install-only.

    Appreciate the feedback and best wishes,
    Erik

Viewing 3 posts - 1 through 3 (of 3 total)
  • You must be logged in to reply to this topic.