Load Balanced Parallel Node Generation for Meshless Numerical Methods

J. Vehovar 1 2 , M. Rot 1 2 , M. Depolli 2 , G. Kosec 2 · 2026

Read Load Balanced Parallel Node Generation for Meshless Numerical Methods by J. Vehovar 1 2 , M. Rot 1 2 , M. Depolli 2 , G. Kosec 2 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-16347.txt.

Description

Jon Vehovar leads three coauthors on parallel node generation for meshless methods. Meshless methods solve partial differential equations by approximating each node's operator as a weighted sum of neighbours. An n-dimensional Poisson disc sampling method generates suitable nodes, handles complex shapes, and supports variable node density.

The authors adapt that method for parallel execution with spatial indexing and work hypertrees. The work hypertree is prebuilt from node density, so each leaf stands for a balanced unit of work. Threads advance separate fronts and claim leaves as needed, avoiding leaves beside those claimed by other threads. Combining node placement constraints with the partly prebuilt hypertree removes the need to lock the tree while it changes. Handling collisions at leaf level sharply cuts the mutex acquisitions needed for point insertion checks. The paper tests the algorithm, compares it with earlier parallelization attempts, and weighs adapting it to distributed systems.

Catalog record

CREATORJ. Vehovar 1 2 , M. Rot 1 2 , M. Depolli 2 , G. Kosec 2
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE26 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-16347

Browse the full library · About the Tower of Babel