Hacker News Reader

Comment by steve1977 | Hacker News Reader

steve1977

I know that I'm simplifying (probably too much), but it seems like things were fine when GitHub was still a Ruby on Rails monolith and all the rigmarole with microservices etc. only made things worse.

View on HN

Unless everything else stays the same (underlying traffic etc.) then you can't really compare. Could be that you hit some fundamental scaling limit with the old design and it completely falls over after a certain scale.

Oh as said I'm pretty sure things are more complex. It's just funny in a way that all these technologies that are usually being sold as "enablers for scale" don't seem to do their job very well.

This sounds more like a belief, based on little more than "correlation is causation", than analysis that controls for macro-trends backed by evidence.

It is, but everyone is entitled to beliefs. Anecdotally, I feel the same way. Everywhere I’ve been, there has always been a legacy monolith that was stable as a rock, with dozens of new microservices scattered around it in an attempt to exit the monolith. The microservices have never once been stable. People fail to take the most basic things into consideration, like “you can’t have Consistency and Availability when everything is a network call.”

I’m sure survivor bias is at play here, but when I look through the older code bases - especially the data model - it’s an entirely different world than the newer stuff, and it’s clear which of the two was written by people who understand systems.

I feel like this accurately described my last two jobs - the rock solid old system that never fails plus a bunch of inconsequential microservices hanging off it (because microservice are the hot new thing, but the microservice people are still smart enough to not replace core functionality from the old system).

I mean one of the selling points of microservices is, that the developer can only be concerned with their slice of the whole and does not need to understand systems. Maybe this is not so clever after all.

Or, it would require an architect who has a very good understanding of the system. Which in reality seems to be rare.

GitHub been oscillating between long phases of "Never any new features but rock-solid and no downtime" and "New features every week but also unicorns (used to be the "service unavailable page") every week" for as long as I can remember. Seems they're on some interval switching between the two.

Reply on news.ycombinator.com