Bully algorithm¶

A distributed leader-election algorithm in which the highest-identifier live process eventually becomes coordinator after failures.

Version: v2 · 2026-09-19 · History
Domain-specific #: 3555
Domain group: Applied Sciences & Engineering
Origin domain: Computer Science & Software Engineering
Subdomain: Distributed Systems → Computer Science & Software Engineering

Core Idea¶

On detecting coordinator failure, a process challenges all higher identifiers; if none responds it announces leadership, otherwise a higher process continues the election. Identifier ordering directs competition upward, timeouts approximate failure detection, and a recovered high-ID process can trigger a new election. The abstraction is therefore identified by a declared carrier, a transformation or constraint over that carrier, and an invariant that tells an analyst whether the named structure is genuinely present.

The load-bearing residual is not the broad topic of distributed systems. It is the domain-specific identity determined by under the synchronous timing and reliable-communication assumptions the highest nonfailed process is eventually uniquely announced coordinator.

How would you explain it like I'm…

Biggest Number Is Boss

Imagine a group of kids who each have a number, and the kid with the biggest number is the leader. If the leader goes missing, a kid calls out to everyone with a bigger number, 'Are you there?' If nobody answers, that kid becomes the leader and tells everyone. If a bigger kid answers, that bigger kid takes over the job of finding the leader.

Picking a New Computer Boss

In a network of computers working together, one computer is often the coordinator, the boss. Each computer has an ID number, and the bully algorithm makes the one with the highest working number the boss. When a computer notices the boss has stopped responding, it sends a message to every computer with a higher number. If none answer within a set time, it announces itself as the new boss. If one does answer, that higher computer continues the election instead. If a computer with a high number comes back after being broken, it can start a new election and take over.

Highest-ID Leader Election

The bully algorithm is a leader-election method in distributed systems. Each process has a unique identifier, and the goal is for the highest-numbered working process to become coordinator. When a process detects that the coordinator has failed, it sends an election message to all processes with higher identifiers. If none of them responds within a timeout, it declares itself coordinator and announces it; if a higher process responds, that process takes over the election. Timeouts are used as a stand-in for detecting failures, and a high-ID process that recovers can start a new election. The guarantee, that exactly one highest working process ends up announced as coordinator, depends on assumptions of synchronous timing and reliable communication.

The bully algorithm is a leader-election protocol for distributed systems in which processes have totally ordered identifiers. On detecting coordinator failure, a process sends election challenges to all processes with higher identifiers; if none responds within a timeout, it announces itself coordinator to the others, and otherwise a responding higher process continues the election. Identifier ordering directs the competition upward, so the process with the highest identifier among the nonfailed processes wins. Failure detection is approximated by timeouts, and a recovered process with a high identifier can trigger a fresh election and displace the incumbent. The correctness guarantee is that, under synchronous timing and reliable-communication assumptions, the highest nonfailed process is eventually announced as the unique coordinator. Without those assumptions, for example with unbounded delays that make timeouts unreliable, the guarantee does not follow.

Scope of Application¶

Bully algorithm belongs to distributed systems and is useful where the analyst can specify the typed distributed systems carrier, defining objects and relations, parameters, conventions, evidence, boundary cases and comparison targets, then evaluate under the synchronous timing and reliable-communication assumptions the highest nonfailed process is eventually uniquely announced coordinator. The scope is broad within that domain but bounded by the need for under the synchronous timing and reliable-communication assumptions the highest nonfailed process is eventually uniquely announced coordinator. Conceptual fault-tolerance algorithm only; no offensive or unauthorized network guidance is provided.

Clarity¶

The abstraction clarifies a crowded vocabulary by making under the synchronous timing and reliable-communication assumptions the highest nonfailed process is eventually uniquely announced coordinator the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test. A bare label is insufficient because the name Bully algorithm can be used for a formal identity, an implementation, or a neighboring result unless carrier and convention are stated.

Manages Complexity¶

Without the abstraction, an analyst must reason directly over many local details: the carrier roles, admissibility assumptions, competing conventions, derived invariants, boundary cases, and proof or validation obligations specific to Bully algorithm. Bully algorithm compresses them into the roles in the structural signature. That compression permits comparison across instances without erasing the variables that determine validity. It also exposes which details may be varied safely and which are constitutive.

Abstract Reasoning¶

Identify the carrier. State what the elements, states, objects, or observations are: the typed distributed systems carrier, defining objects and relations, parameters, conventions, evidence, boundary cases and comparison targets. Reject examples whose alleged carrier belongs to a different problem.
Lock the constitutive rule. Express under the synchronous timing and reliable-communication assumptions the highest nonfailed process is eventually uniquely announced coordinator independently of one notation or implementation.

Knowledge Transfer¶

Knowledge transfers strongly among subfields of distributed systems because they reuse the typed distributed systems carrier, defining objects and relations, parameters, conventions, evidence, boundary cases and comparison targets, Identifier ordering directs competition upward, timeouts approximate failure detection, and a recovered high-ID process can trigger a new election., and type the carrier, state every parameter and convention in the definition, test that under the synchronous timing and reliable-communication assumptions the highest nonfailed process is eventually uniquely announced coordinator, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases.

Relationships to Other Abstractions¶

Current abstraction Bully algorithm Domain-specific

Parents (1) — outgoing typed relations

Bully algorithm is a kind of Coordination Prime

What makes it its own entry: the domain-specific identity determined by under the synchronous timing and reliable-communication assumptions the highest nonfailed process is eventually uniquely announced coordinator.

Hierarchy paths (5) — routes to 4 parentless roots

Bully algorithm → is a kind of Coordination → presupposes Concurrency

Show alternative paths (4)

Neighborhood in Abstraction Space¶

Bully algorithm sits in a crowded region of the domain-specific corpus (40^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Algorithmic Coordination & Fair Allocation (15 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-25