We will take a look at some of the tools used for monitoring and issue diagnosis in a large scale PAAS hosting platform, and then take a look at how we can dig deeper to resolve performance issues using Blackfire.

Comments

Comments are closed.

Rob Wilson at 21:00 on 8 May 2019

Fantastic talk by Nick, some great information passed across. Very brave in doing the live demo (crazy ;) ). A great way of getting around the live demo is to use a video (then you can talk over it whilst not focusing on typing and screen switching)

Lucia Velasco at 21:04 on 8 May 2019

I liked the helicopter graph! Also the xkcd chart.
Loved the real (recognisable!) examples of sudden high traffic.
I liked that you tied together the different monitoring screens to highlight the correlation between data.
As a front end dev... Sorry! ?
Great breakdown of profiling and I enjoyed the discussion on how people think about profiling (i.e. when people think it's not easyable enough).
I think it's just because I'm low on sleep, but the dark background made me a bit sleepy.
I liked the live demo, it was helpful seeing how the graph held you highlight possible pain points.
Thank you so much!

Ben Kelly at 21:05 on 8 May 2019

Really enjoyed the live demo. We will definitely be trying Backfire as a direct result!

Sam Rogers at 21:08 on 8 May 2019

Very well-structured and thought out talk from Nick, he clearly knows these tools very well. I especially enjoyed the visualisation and the live demo (extremely brave!)

It was a little hard to hear, more in the beginning I think

Nock rolled with the punches very well, the improved code being the wrong way round was quickly diagnosed and it didn't cause any issues.