Thank you ! This was a pretty good start to understand K6. Keep this objective narrative in your videos. Good job!

@TonyTeachesTech

2 жыл бұрын

Glad you enjoyed it!

Hi Tony this is extremely good we want more video series on this so we learn and enjoy it...

Love this. So clear Thank you

@TonyTeachesTech

2 жыл бұрын

You're welcome!

That was great and simple, thank you

@TonyTeachesTech

3 жыл бұрын

Glad it was helpful!

Great job

Clear and simple!

@TonyTeachesTech

3 жыл бұрын

Thanks!!

Thanks man :D

Thanks for the video. very insightful. :)

@TonyTeachesTech

3 жыл бұрын

Glad it was helpful!

Thank you very much. This is very helpfull and exciting.

@TonyTeachesTech

2 жыл бұрын

Glad it was helpful!

Nicely Explained!

@TonyTeachesTech

2 жыл бұрын

Thanks!

Thanks for this content

@TonyTeachesTech

2 жыл бұрын

You bet

that was good of explanation and very help for us 'Thank you'

@TonyTeachesTech

2 жыл бұрын

You are welcome!

Great video.

@TonyTeachesTech

3 жыл бұрын

Thanks!

Thank you so much!

@TonyTeachesTech

2 жыл бұрын

You're welcome!

Thank you.

@TonyTeachesTech

Жыл бұрын

You're welcome!

This was a very nice introduction to the tool. Thanks! If I can have a constructive comment: when analyzing performance (total network time, latency, ...) you probably are trying to understanding the service quality and checking if SLAs are being meet (or not). In this context the "avg" is not the best indicator (almost never), give as it's an average it's not representing any specific request a "consumer" got. Then for instance, you might have a "good avg time" (109ms) while 50 clients waited for 1000ms and 450 clients waited for 10ms. That could mean 50 clients are under the SLA. TL; DR using percentiles is more representative, and enables you to understand "group" of requests; also taking a look to max/ min is nice to understand your worst case, and so on.

@TonyTeachesTech

3 жыл бұрын

Very good point! I'll definitely be using percentiles in similar cases in the future. Thank you 🙏

Good job, great explation of k6. Please keep the great content coming. How about some video content on Wordpress hardening and security. Once again thanks.

@TonyTeachesTech

3 жыл бұрын

Coming soon!

Nice bro 👍

@TonyTeachesTech

3 жыл бұрын

Thanks ✌

Hi @Tony Nice informative video on K6. What is your opinion on using Locust for load testing compare to K6 tool? Also can you please help me to know, if I want to send 10% illegal traffic and 90% of legal traffic to my application how I can achieve the same from K6 java script test file? Thanks

The video was very informative, can you by a request make a video so we can understand how we can represent the test results using influx 2.0 as database and grafana, Thank you.

What about package k6 + kafka plugin in Container?

Please make a video about HTTP Strict Transport Security

@TonyTeachesTech

3 жыл бұрын

You got it. Coming up

Hi Tonny, Can you make a woocommerce checkout load test similuations? Like starting from the landing page then adding to cart checkout filling information load test. I couldn't find it on the internet for woocommerce. I don't even say on youtube anything. thanks

@TonyTeachesTech

3 жыл бұрын

Yes I believe you can use the k6 browser recorder to mimic the behavior of the user k6.io/docs/test-authoring/recording-a-session/browser-recorder I'm not 100% sure how this works with stuff in the cart (like with cookies and server sessions)

Can this be used to determine what level of vps(Eg in terms of ram,cpu) you need to run a wordpress site ? If not what can you use to estimate server specs for wordpress?

@TonyTeachesTech

3 жыл бұрын

Load testing like this could help you figure out if you need to increase your server resources, but it won't quantitatively tell you how much you need. I'm not sure of a tool that outputs server specs. If you're on VPS hosting and see page load times slowing down at times of peak traffic, that's an indicator to scale up your resources. Another option is to host with a company like Kinsta that is based on monthly page views and automatically scales up its resources when your website sees spikes in traffic.

@alldayieatlikeashark

3 жыл бұрын

@@TonyTeachesTech makes sense, thanks!

I think a clarification is in order on metrics. The duration is important but the average was focused on. This would be bad. Far too often people focus on averages but this can land you in trouble. Assuming your txn times are normally distributed then that means half of the times are slower than your mean. Not good. So step back and think what is it you want to learn and convey after performing a load test? Typically you want to convey: 1. How is it performing in general 2. Are there any problems (or potential problems)? Using a mean seams like a good way to answer number 1 but it is not. The P95 is much better. For item 2, you need to look at that P95 number and see what is happening to give you a number that is not "perfect". Granted we cannot guarantee 100% speedy txns in the real world but what are those 5% doing? Are you still meeting your SLO/SLA numbers? Are these worrisome numbers where you are seeing a bunch of timeouts but they are hidden by the volume of acceptable numbers? This is where the rubber meats the road for load testing. You have to be able to see what is going on and decide if it meets expectations and while you may be tempted to use an easier number (mean) to do so, you really shouldn't.

@TonyTeachesTech

3 жыл бұрын

I appreciate this perspective. I will try to better understand p95 and communicate the importance to my audience in future videos. Thanks Nick

@NickFrom1228

3 жыл бұрын

@@TonyTeachesTech No problem. Here's one really good video to provide a frame of reference. Enjoy. kzread.info/dash/bejne/nn5s28aCpbOslbg.html&ab_channel=StrangeLoopConference

@TonyTeachesTech

3 жыл бұрын

@@NickFrom1228 Thank you Nick :)

0:20 actually that's oo-boon-too ;)

interesting.. it does not give the cpu and memory stats

How many of you came here after watching hitesh's k6 testing 1min video : )

@TonyTeachesTech

3 жыл бұрын

What's that?