search

mikhailshilkov / mikhailio-hugo

Sources of https://mikhail.io migrated to Hugo
MIT License
12 stars 8 forks source link

Comments to "Visualizing Cold Starts" #6

Open mikhailshilkov opened 5 years ago

mikhailshilkov commented 5 years ago

Add your comment to Visualizing Cold Starts. The comments will be displayed directly on the page.

JRiggenbach commented 5 years ago

Nice way to visualize the startup times. The way you deploy your application, especially when it comes to node functions, can greatly increase increase your cold startup times. How were each of these applications deployed?

johnz commented 5 years ago

Are you using windows or Linux instance for Azure test, can it make any difference?

mikhailshilkov commented 5 years ago

@JRiggenbach On Azure I deploy with "Run from Package" with a local zip file, that's currently the recommended way from the team. AWS and GCP are deployed with a zip sitting in S3/Cloud Storage, both controlled from Pulumi. Which deployment options are you referring to as greatly increasing the startup time? Any links?

mikhailshilkov commented 5 years ago

@johnz Azure test is on Windows. Currently, only python runs on Linux on Consumption plan, and that's still in preview. You can check the comparison in Azure Functions on Linux.

ThomasKeralla commented 4 years ago

Hi Mikhail. Very interesting read, and love your content in general. I am writing my thesis right now on FaaS and was wondering how you track function instances? how do you get the id?

mikhailshilkov commented 4 years ago

Thanks for the kind words! You can find the ID extraction in the code:

ThomasKeralla commented 4 years ago

Thanks for your reply. Unfortunately it dose not seem to be a solution with python, we can only get to what seems to be the python process running in the app and not the app itself. But thanks for sharing :)