Distribution
Wear / Endurance Distribution (DoEC)
What it shows. Left: a Lorenz-style curve of erase/reset counts (units sorted hottest-first) — a flat curve means even wear, a steep one means a few units absorb most of the wear. Right: the Degree of Erase Count (DoEC, standard deviation of erases per unit) for each engine.
Why it matters. SSD lifetime is bounded by per-block erase endurance, so even wear (low DoEC) extends device life. This is the endurance counterpart to WAF: a policy can achieve low WAF yet concentrate wear on a few blocks. Comparing engines here shows the wear-levelling side of each placement strategy.
What we found
- Coarse erase concentrates wear. On prxy_0, superblock GC nearly doubles the wear spread vs block GC (DoEC 20.2 vs 10.6) — the endurance price of the lower-WAF-looking coarse reclaim.
- WAF and wear can diverge. ZNS hot/cold-greedy on prxy_0 keeps DoEC low (10.4) despite a high WAF, whereas FDP-PI trades a little wear spread (12.4) for much lower WAF.
- Wear-count distributions are mostly uni-modal (Sarle < 0.55) except the pathological coarse-GC cases, i.e. wear is spread rather than split into hot/cold block populations.
prxy_0 (MSR real)
prxy_1 (MSR real)
prxy_1 (3M writes, real)
TPC-C (W2K8, real)
uniform (synthetic)
Zipf-1.2 (synthetic)
80/20 (synthetic)
sequential (synthetic)