Skip to content
Free public beta until December 4, 2026: every course is open, no card needed. After that, the first chapter of each course stays free and full access is a one-time $39 ($19 for up to 100 founding members chosen for helpful feedback). Your progress is kept. How the beta works

data reduction

Computing on the device so that much less data has to leave it, such as a few features instead of every sample.

Data reduction is processing at or near the source that replaces a high-rate raw stream with smaller derived data (features, summaries or events) before transmission, trading the raw samples for lower bandwidth, power and exposure.

Examples

  • Sending one 24-byte feature vector a second instead of 600 bytes of raw accelerometer samples cuts the fan's daily traffic from 60 MB to about 10 MB.

  • Data reduction throws the raw samples away for good, so keep a short raw capture available on command.

See also

feature vector edge computing batching
Read the full section →