DeepAI AI Chat
Log In Sign Up

Derivation of Heard-Of Predicates From Elementary Behavioral Patterns

by   Adam Shimi, et al.

There are many models of distributed computing, and no unifying mathematical framework for considering them all. One way to sidestep this issue is to start with simple communication and fault models, and use them as building blocks to derive the complex models studied in the field. We thus define operations like union, succession or repetition, which makes it easier to build complex models from simple ones while retaining expressivity. To formalize this approach, we abstract away the complex models and operations in the Heard-Of model. This model relies on (possibly asynchronous) rounds; sequence of digraphs, one for each round, capture which messages sent at a given round are received before the receiver goes to the next round. A set of sequences, called a heard-of predicate,defines the legal communication behaviors – that is to say, a model of communication. Because the proposed operations behave well with this transformation of operational models into heard-of predicates, we can derive bounds, characterizations, and implementations of the heard-of predicates for the constructions.


page 1

page 2

page 3

page 4


Characterization and Derivation of Heard-Of Predicates for Asynchronous Message-Passing Models

Message-passing models of distributed computing vary along numerous dime...

K set-agreement bounds in round-based models through combinatorial topology

Round-based models are the main message-passing models; combinatorial to...

Characterizing Asynchronous Message-Passing Models Through Rounds

Message-passing models of distributed computing vary along numerous dime...

Message Reduction in the Local Model is a Free Lunch

A new spanner construction algorithm is presented, working under the LOC...

Fault-Tolerant Graph Realizations in the Congested Clique

In this paper, we study the graph realization problem in the Congested C...

Computing Power of Hybrid Models in Synchronous Networks

During the last two decades, a small set of distributed computing models...