Design interviews / Intermediate / 5 min
Design Netflix: adaptive bitrate streaming
The player picks 240p to 4K for every 4-second segment. Ride a train through tunnels and compare fixed, speed-based and buffer-based choices.
Start the simulation → Change the load, break a part, watch what happens. Free, no sign-up.
In short
- What it is
- Adaptive bitrate streaming stores each film in several qualities cut into short segments, and the player picks one per segment as it goes.
- Why it matters
- Viewers have very different and changing connections, and stalls make people stop watching far more than a softer picture.
- The trade-off
- Higher quality looks better but risks stalls; switching often keeps playback smooth but the picture visibly jumps.
- Say it in an interview
- "HLS/DASH manifests with a bitrate ladder per title, segments on a CDN (Netflix Open Connect), and a hybrid ABR on throughput and buffer level."
Check yourself
Why does always streaming 4K on an 8 Mbps connection keep stopping?
Answer: Each segment takes longer to download than to play, so the buffer runs empty. 15 Mbps of video over 8 Mbps means 4 seconds of film take 7.5 seconds to fetch.
What does the video buffer protect against?
Answer: Short drops in network speed, by playing video already downloaded. With 20 to 30 seconds buffered, a tunnel shorter than that causes no stall.
What is a weakness of choosing quality only by measured speed?
Answer: It reacts after the speed has already dropped, so a sudden dip can still empty the buffer. Throughput estimates lag; buffer-based rules (and hybrids like Netflix uses) also watch how much video is left.