Skip to content

Check Timing, Cost and Fleet Evidence

Knowledge check · self-test

Check Timing, Cost and Fleet Evidence

Identify where an edge pipeline records a local decision, its evidence version, and its outcome for later review.

Back to the chapter →

Identify where an edge pipeline records a local decision, its evidence version, and its outcome for later review.

## {#depth-boundary-edge-patterns-processing-1-open .unlisted .unnumbered .depth-l0 .depth-boundary} ::: {.callout-tip} ## Data Dora, Data and analytics domain expertCheckpoint: Price the Trade-off {.depth-l0} - Edge ML is attractive for 10-50 ms decisions, while cloud ML often pays a 200-500 ms round trip. - In this illustrative comparison, batch processing reduces compute cycles by 80-95%, but introduces 1-60 minute detection latency. - Cost savings need enough volume: the chapter's decision framework separates small deployments, medium deployments, and large deployments above 10,000 sensors or 10 GB/month. ::: ## {#depth-boundary-edge-patterns-processing-1-close .unlisted .unnumbered .depth-l2 .depth-boundary} ## {#depth-boundary-edge-patterns-processing-2-open .unlisted .unnumbered .depth-l0 .depth-boundary} ::: {.callout-tip} ## Data Dora, Data and analytics domain expertCheckpoint: Validate the Fleet Mix {.depth-l0} - The illustrative supermarket fleet starts with 40,000 stores, 12 display cases per store, and 3 sensors per case. - The raw daily total is 33.2 GB; the illustrative mixed edge plan models 1,445.38 MB after processing, with compressor byte savings depending on summary encoding. - Under the stated decimal MB and EUR 0.50/MB assumptions, modeled daily cellular cost drops from EUR 16,600 to about EUR 722.69; each sensor type uses the pattern that matches its evidence need. ::: ## {#depth-boundary-edge-patterns-processing-2-close .unlisted .unnumbered .depth-l2 .depth-boundary}

Back to Edge Processing: Aggregation Decisions: Keep Transformation Separate from Storage