Towards a Linear-Algebraic Hypervisor

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Considine, Breandan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917408358268928
author Considine, Breandan
author_facet Considine, Breandan
contents Many techniques in program synthesis, superoptimization, and array programming require parallel rollouts of general-purpose programs. GPUs, while capable targets for domain-specific parallelism, are traditionally underutilized by such workloads. Motivated by this opportunity, we introduce a pleasingly parallel virtual machine and benchmark its performance by evaluating millions of concurrent array programs, observing speedups up to $147\times$ relative to serial evaluation.
format Preprint
id arxiv_https___arxiv_org_abs_2604_12902
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Towards a Linear-Algebraic Hypervisor
Considine, Breandan
Programming Languages
Distributed, Parallel, and Cluster Computing
Performance
Many techniques in program synthesis, superoptimization, and array programming require parallel rollouts of general-purpose programs. GPUs, while capable targets for domain-specific parallelism, are traditionally underutilized by such workloads. Motivated by this opportunity, we introduce a pleasingly parallel virtual machine and benchmark its performance by evaluating millions of concurrent array programs, observing speedups up to $147\times$ relative to serial evaluation.
title Towards a Linear-Algebraic Hypervisor
topic Programming Languages
Distributed, Parallel, and Cluster Computing
Performance
url https://arxiv.org/abs/2604.12902