The misses own the mean

How it behaves

One input on the horizontal axis, the others held fixed.

Worked examples, recomputed

Every number the mechanisms quote for this relationship is recomputed from it, so none of them disagrees with it.

• 40% in front of a 50 ms store → 19.96 ms saved
• 95% in front of a 0.2 ms probe → 0.19 ms saved
• pooling with a slow origin gains ≈ 7.2 ms
• remote round trips cost ≈ 0.43 ms
• with a 2 ms origin the gain is ≈ 0.22 ms
• 99% cost 1, 1% cost 500 → mean 5.99, the tail is 83% of it
• quarantining the 1% of ₹1,950 orders hides 4.2% of revenue (true average ₹465)
• 990 at 5 ms, 10 at 700 ms → mean 11.95 ms

Where it appears

Each mechanism states it in its own names. The general symbol each name stands for is shown with it.

• Caching

  What happens when we deliberately keep a second copy of data that no protocol forces to be correct?

  $$h\cdot t_{\text{hit}}+(1-h)\cdot t_{\text{miss}}$$

  Read it in place →