News for hackers. Stuff that matters.

A comment on Serverless: slower and more expensive. Parent comment

Re:Serverless: slower and more expensive by unicornmama on Monday September 23, 2019 @10:24AM (#21049187)
I will argue the opposite. Startups take on enough risks as it is. Unless your startup requires or is about a novel architecture. Why add more risks with non battle hardened technology.

Software professionals often sees benefits without understanding the tradeoffs.

[ Parent | on HN ]
Re:Serverless: slower and more expensive by abiro on Monday September 23, 2019 @11:30AM (#21049834)
Lambda is 5 year old technology. This is like arguing in 2011 that startups shouldn’t use EC2, because it’s “risky”.
[ Parent | on HN ]
Re:Serverless: slower and more expensive by unicornmama on Monday September 23, 2019 @01:12PM (#21050960)
The technology age isn’t the issue. The issue is how many projects have successfully deployed large scale reliable systems built with Lambda.
[ Parent | on HN ]
Re:Serverless: slower and more expensive by abiro on Monday September 23, 2019 @03:55PM (#21052898)
The internet is full of success stories if you care to look. My favorites:

- iRobot (maker of Roomba) has been running its entire IoT stack serverless since 2016 (architect Ben Kehoe is a worthwhile follow on Twitter)

- Reddit’s video hosting service is built and operated by a single engineer on a serverless stack

[ Parent | on HN ]
Re:Serverless: slower and more expensive by danillonunes on Tuesday September 24, 2019 @11:45PM (#21067410)
Reddit’s self hosted video is terrible. Using them to advocate serverless is like using Twitter on the fail whale days to advocate Rails.
[ Parent | on HN ]
Re:Serverless: slower and more expensive by unicornmama on Tuesday September 24, 2019 @04:25AM (#21057733)
Thanks. This is good info.
[ Parent | on HN ]
Re:Serverless: slower and more expensive by laurentl on Tuesday September 24, 2019 @03:14AM (#21057363)
Large scale reliable systems are antinomic with “launching a startup”. You’re going to go through 2 or 3 pivots and as many refactors, large scale is the last thing you want to optimize for.
[ Parent | on HN ]