Serverless Computing: One Step Forward, Two Steps Back

A few weeks back, seven researchers at UC Berkley published an article panning serverless computing in general, by the title of this thread: “Serverless Computing: One Step Forward, Two Steps Back”
The article can be found here:


or more specifically here: https://arxiv.org/pdf/1812.03651.pdf

I’ve been searching for responses from the serverless community, but haven’t found much of anything so far. So, I thought I’d start a thread to foster some constructive feedback.

Yan Cui addressed this whitepaper in his article You are thinking about serverless costs all wrong.

I wonder what other serverless thought leaders think about this paper and mainly issues listed there:

  1. Limited Lifetimes
  2. I/O Bottlenecks
  3. Communication Through Slow Storage
  4. No Specialized Hardware