0-7-1-understand-jitter

name	0.7.1-understand-jitter
description	[0.7.1] Randomized variance introduced into the computed delay to decorrelate competing threads and mitigate thunderin

understand-jitter

CALL NUMBER: deep_spinlock.jitter : deep_exponential_backoff(2) DEFINITION: Randomized variance introduced into the computed delay to decorrelate competing threads and mitigate thundering-herd synchronization on lock release.

Invoke this skill to understand jitter down to its primitives. The RELATIVE ROOT below is the least-fixed-point closure of everything it bundles from — the full import cone, grouped by the lib each prim comes from. Projected from a prover-typed KB (MAP/SWI-Prolog consistency gate): every reference below resolves.

THE RELATIVE ROOT (the import cone, by lib)

from deep_exponential_backoff

• djt_random_uniform_sample (d1): The uniform random draw within the bounding range that produces the new current_delay value; the randomness decorrelates this thread's retry timing from concurrent threads.
• djt_wave_attenuation (d2): The emergent property that decorrelated jitter reduces the probability of simultaneous retries among competing threads; achieved through per-thread history-dependent ranges that diverge over time.

from deep_spinlock

• decorrelated_jitter (d1): Jitter strategy where each thread computes delay = random.uniform(min_delay, previous_delay * 3); history-dependent decorrelation reduces coordinated retry waves.
• equal_jitter (d1): Jitter strategy computing delay = base_delay / 2 + random.uniform(0, base_delay / 2); ensures wait never falls below half the nominal delay.
• full_jitter (d1): Jitter strategy selecting a uniform random value in the range [0, current_delay]; maximizes desynchronization at the cost of potentially very short waits.

CONSUMERS (what needs this)

exponential_backoff

---

Projected from the concurrency synchronization primitives and memory models KB (407 concepts / 477 relations) — consistency-typed by MAP; the facet list after the colon IS the cross-lib dependency web.

(leaf — this is an actual skill.)