Synthetic examples / no hardware measurements

SSD steady-state window counterexamples

Name: SSD steady-state window counterexamples
Creator: Hesela
Published: 2026-10-12
License: https://creativecommons.org/publicdomain/zero/1.0/

Six synthetic five-round IOPS examples separate variation, drift and local stability. These are arithmetic counterexamples, not SSD benchmark measurements.

Download CSV6 cases / 12 columns

JSON and provenance / Evaluator code / Research explanation

Method

Each row contains five nonnegative invented IOPS values at equally spaced round indices 0-4. Ordinary least squares yields a fitted slope; its absolute value times four gives fitted span. Both observed range and fitted span are divided by the five-value arithmetic mean. Inclusive 20% and 10% limits illustrate SNIA SSS PTS 2.0.2 clause 2.1.24 only. Missing values are rejected. All-zero windows are rejected because normalization is undefined.

Invented IOPS values. R = range / mean; T = fitted span / mean. Display rounded to one decimal; decisions use full precision.
Scenario	Rounds 1-5 (IOPS)	R	T	Both checks
small-variation	100, 102, 99, 101, 98	4.0%	2.0%	Pass
gradual-drift	92, 96, 100, 104, 108	16.0%	16.0%	Fail
symmetric-swings	88, 112, 100, 112, 88	24.0%	0.0%	Fail
early-plateau	200, 200, 200, 200, 200	0.0%	0.0%	Pass
later-plateau	100, 100, 100, 100, 100	0.0%	0.0%	Pass
asymmetric-excursion	95, 95, 115, 95, 100	20.0%	4.0%	Pass

Interpretation

Gradual drift passes the range check but fails the trend check. Symmetric swings do the reverse. The asymmetric case passes even though a point exceeds 110% of its mean: range and maximum deviation from the mean are different. Two individually passing plateaus can still differ by a factor of two.

Reproduce

The evaluator performs no device I/O. Download the linked module, then run this command in its directory with Node.js. No third-party packages are required. It prints the exact CSV; provenance JSON records hashes of both CSV and module.

node --input-type=module -e 'import { toCsv } from "./ssd-steady-state.mjs"; process.stdout.write(toCsv())'

Limits

No real device, observation period, sample population, random seed or confidence interval.
This numerical illustration does not implement preparation, workload matrices, all tracked variables or full PTS reporting.
A passing window cannot establish future stability, statistical stationarity, acceptable speed or tail latency.
Rounds are equally spaced test indices, not necessarily seconds. Do not pass irregular time samples to this API.
Floating-point comparisons allow a 64-machine-epsilon tolerance at each threshold.
Early and later plateaus can be concatenated to demonstrate a drop between two individually passing windows.

Column definitions

scenario: Synthetic case identifier; not a device model.
round1: Invented IOPS in round 1.
round2: Invented IOPS in round 2.
round3: Invented IOPS in round 3.
round4: Invented IOPS in round 4.
round5: Invented IOPS in round 5.
mean: Arithmetic mean IOPS of the five values.
rangePercent: 100 times (maximum minus minimum) divided by mean.
fittedSpanPercent: 100 times absolute least-squares slope times four round intervals divided by mean.
rangePass: Range at most 20% of mean.
trendPass: Fitted span at most 10% of mean.
pass: Both numerical checks pass; not PTS certification.

Sources and license

CC0-1.0 for Hesela's original code and invented values; source publications retain their rights.

SNIA. Solid State Storage Performance Test Specification, version 2.0.2 (1 October 2020), definitions 2.1.1, 2.1.3, 2.1.6, 2.1.13, 2.1.18 and 2.1.24.

Definitions: SSD steady state, preconditioning and active LBA range.