Retries in distributed services often expose hidden race conditions and state inconsistencies. We present a Go-based engine that enforces idempotent request handling by coupling a synchronized idempotency map with a lightweight resource store. The engine guarantees that each unique idempotency key maps deterministically to a single response, thereby eliminating duplicate side effects even under concurrent retries. We formalize the causal ordering invariant that the engine preserves a strict height-balanced key-ordered structure for the idempotency map, ensuring O(log n) lookup and insertion. Empirical evaluation on an AMD EPYC 7763 host yields a mean latency of 1399.27 µss, a P95 latency of 2287.16 µss, and a throughput of 714,659 ops/s over 30 iterations.
Manuscript Attribution: This manuscript is an empirical research preprint synthesized via agentic autonomous systems engineering pipelines adhering to academic two-column preprint conventions under the direction of Mateus Yonathan.
Hardware & Algorithmic Provenance: Algorithmic complexity bounds, execution profiles, and memory models cite verified static AST analysis and compiler toolchains evaluated by autonomous agent feedback loops.
Limitation of Liability & Scholarly Purpose: Disseminated solely for scholarly inquiry, peer discussion, and reproducible empirical comparison. It does not constitute formal peer-reviewed accreditation. Any reliance upon the mathematical proofs or complexity bounds herein is strictly at the reader's own legal and technical risk.
@article{yonathan2026_2026_10_24_retries_are_a_distributed_systems_b,
title={Retries Are a Distributed Systems Bug Waiting to Happen: Runtime Invariants and Algorithmic Bounds in Go},
author={Yonathan, Mateus},
journal={Polyglot Systems Lab Agentic Research Preprints},
year={2026},
publisher={Polyglot Systems Lab},
url={https://myonathanlinkedin.github.io/papers/2026-10-24_retries_are_a_distributed_systems_b/},
howpublished={\url{https://raw.githubusercontent.com/myonathanlinkedin/mini-papers/main/papers/2026-10-24_retries_are_a_distributed_systems_b/2026-10-24_retries_are_a_distributed_systems_b.pdf}}
}