march 2023: a small team inside Amazon publishes a receipt against the company’s own pitch
nobody noticed the post — until everybody did
what the service does · why the design was textbook · what killed it · and what to steal
video quality analysis: take the real audio and video, run detectors on it, flag what a human would notice
three components, stitched together by a state machine, with S3 as the hand-off in the middle
all the hard distributed-systems machinery — queueing, retries, state — rented, not built
the orchestrator meters every step it takes — and this pipeline never stops stepping
prime video slammed into AWS throttles at roughly five percent of expected load
one second of video becomes a pile of frame objects — written once, read back by everyone
durability priced for forever — bought for data that dies in a second
the same frame, handed over two different ways — same work, different physics
converter and detectors compiled into one service — the state machine demoted to control flow
you don’t break the monolith apart to scale it — you clone it
one deployable unit — scaling by replication, specializing by configuration
the rewrite didn’t just cut the bill — it made the original ambition affordable
“if the company selling the dream can’t make it work — why are you paying the tax?”
cockcroft: this was never microservices-to-monolith — it was the playbook, working
the referee’s ruling: when scale or economics change, you change the design
for every arrow between two boxes: how big is the payload, and how often does it cross?
you split services where teams split — not where the whiteboard looks tidier
the bill is lower — the operational burden is back on the team
what to take straight into your own systems
one clean chain — from textbook pipeline to a ninety-percent receipt