Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 0707.3435

A Knowledge-Based Analysis of Global Function Computation

Joseph Y. Halpern·Sabina Petride
View original source ↗Read source document ↗Export metadata ↓

Abstract

Consider a distributed system N in which each agent has an input value and each communication link has a weight. Given a global function, that is, a function f whose value depends on the whole network, the goal is for every agent to eventually compute the value f(N). We call this problem global function computation. Various solutions for instances of this problem, such as Boolean function computation, leader election, (minimum) spanning tree construction, and network determination, have been proposed, each under particular assumptions about what processors know about the system and how this knowledge can be acquired. We give a necessary and sufficient condition for the problem to be solvable that generalizes a number of well-known results. We then provide a knowledge-based (kb) program (like those of Fagin, Halpern, Moses, and Vardi) that solves global function computation whenever possible. Finally, we improve the message overhead inherent in our initial kb program by giving a counterfactual belief-based program that also solves the global function computation whenever possible, but where agents send messages only when they believe it is necessary to do so. The latter program is shown to be implemented by a number of well-known algorithms for solving leader election.

Explore related subjects

cs.DCcs.LO

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Joseph Y. Halpern, Sabina Petride. 2007-07-23. A Knowledge-Based Analysis of Global Function Computation. https://arxiv.org/abs/0707.3435

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Deterministic Self-Stabilizing BFS Construction in Constant Space

In this paper, we resolve a long-standing question in self-stabilization by demonstrating that it is indeed possible to construct a spanning tree in a semi-uniform network using constant memory per node. We introduce a self-stabilizing synchronous algorithm that builds a breadth-first search (BFS) spanning tree with only $O(1)$ bits of memory per node, converging in $2^ε$ time units, where $ε$ denotes the eccentricity of the distinguish node. Crucially, our approach operates without any prior knowledge of global network parameters such as maximum degree, diameter, or total node count. In contrast to traditional self-stabilizing methods, such as pointer-to-neighbor communication or distance-to-root computation, that are unsuitable under strict memory constraints, our solution employs an innovative constant-space token dissemination mechanism. This mechanism effectively eliminates cycles and rectifies deviations in the BFS structure, ensuring both correctness and memory efficiency. The proposed algorithm not only meets the stringent requirements of memory-constrained distributed systems but also opens new avenues for research in self-stabilizing protocols under severe resource limitations.

cs.DC↗
arXiv2026

Space Filling Curves is All You Need: Communication-Avoiding Matrix Multiplication Made Simple

General Matrix Multiplication (GEMM) is the cornerstone of HPC workloads and Deep Learning. State-of-the-art (SOTA) vendor libraries tune tensor layouts, parallelization schemes and cache blocking to minimize data movement across the memory hierarchy and maximize throughput. However, optimal settings for these parameters depend on the target platform and matrix shapes, making exhaustive tuning infeasible. In this work, we address this cumbersome scheduling search and tuning using space-filling curves (SFC). We partition the matrix multiplication using advancements in SFC, and obtain platform-oblivious and shape-oblivious matrix multiplication schemes with a high degree of data locality. We extend the SFC-based work partitioning to implement Communication-Avoiding (CA) algorithms with replication techniques in a seamless fashion. The resulting SFC-CA GEMM achieves provable asymptotic communication optimality for both square and rectangular matrix regimes. Across four x86 and Arm platforms, SFC-CA GEMM outperforms vendor libraries by up to 5.5$\times$ per shape and 1.8$\times$ in weighted harmonic mean (WHM) throughput. Last, we show the impact of our work on two real-world applications by leveraging our SFC-CA GEMM as a compute backend: i) prefill of LLM inference with speedups up to 1.85$\times$ over SOTA inference runtimes, and ii) distributed-memory matrix multiplication with speedups up to 2.3$\times$ over the SOTA distributed-memory GEMM framework with vendor-optimized compute backend.

cs.DC↗
arXiv2026

MoEless: Efficient MoE LLM Serving with Serverless Experts

Large Language Models (LLMs) increasingly adopt Mixture-of-Experts (MoE) architectures to scale efficiently under stringent resource constraints. However, MoE's sparse activation causes severe expert load imbalance, where a few experts become stragglers while others remain underutilized, leading to inflated inference latency and cost. Existing solutions assume static, serverful model deployments, limiting expert elasticity and often incurring costly expert swapping or degraded output quality. We present MoEless, an efficient serverless MoE serving framework that mitigates expert load imbalance via elastic expert execution. MoEless leverages lightweight, layer-aware predictors to estimate incoming expert load distributions and proactively identify stragglers. We design optimized scaling and placement strategies to improve function locality, GPU utilization, and cross-expert load balance. MoEless is prototyped on top of Megatron-LM and deployed on an eight-GPU testbed. Experiments with open-source MoE models and real-world workloads show that MoEless reduces inference latency by 43% and inference cost by 84% compared to state-of-the-art solutions.

cs.DC↗