CODEWITHSUNDEEP
next.jsjmetersingle-page-application
Daryl Revilla
Daryl Revilla

Reputation: 1

LoadTesting a Single Page Application (SPA)

We wanted to conduct load testing on a web application built-in next.js with integration with looker.

  1. Recorded the scripts in JMeter, requests we get are: .json, and APIs.
  2. Set HTTP Request Defaults > Advanced > checked "Retrieve All Embedded Resources" and "Parallel downloads. Number 6 (emulating Chrome browser)
  3. Ran the a baseline test
  4. Observed that, each steps took roughly rough around 100+ seconds, and found out that it's the .json that are taking too long to response back.

having said this what would be the best course of action?

Upvotes: 0

Views: 301

Answers (2)

Daryl Revilla
Daryl Revilla

Reputation: 1

All - there's a workaround/solution I found from Blazemeter, and it uses a plug-in called Parallel Controller.

https://www.blazemeter.com/blog/parallel-controller-in-jmeter

Which emulates sending HTTP request concurrently similar to LoadRunner's web_concurrent (start/end) function.

Upvotes: 0

Dmitri T
Dmitri T

Reputation: 168072

I don't know what "baseline test" is, normally you should start with 1 thread (virtual user) and gradually increase the load at the same time looking into the following metrics:

  • number of threads
  • response time
  • throughput (number of requests per second)
  • response status codes

Ideally the throughput should grow proportionally to the number of threads and response time should remain the same, but in reality at some point you will see response time going up and/or errors start occurring and throughput will go down. It will indicate the bottleneck in the system under test.

It would be also a good idea to make sure that you're not getting false-negative results because JMeter cannot send requests fast enough, make sure to follow JMeter Best Practices and that JMeter load generator(s) have enough headroom to operate in terms of CPU, RAM, Network, etc.

Upvotes: 0

Related Questions