Reclaiming Idle CPU Cycles on Kubernetes: Sparse-Domain Multiplexing for Concurrent MPI-CFD Simulations

TL;DR

This paper introduces a multiplexing framework for Kubernetes that reclaims idle CPU cycles during MPI-CFD simulations, significantly improving throughput by co-locating multiple simulations with minimal overhead.

Contribution

It presents a novel multiplexing approach with a dynamic controller and analytical model to efficiently run multiple MPI simulations concurrently on shared Kubernetes clusters.

Findings

Achieved 1.77x throughput with two simulations

Scaling to 3.74x throughput with five simulations

Automated pipeline with no manual intervention or pod restarts

Abstract

When MPI-parallel simulations run on shared Kubernetes clusters, conventional CPU scheduling leaves the vast majority of provisioned cycles idle at synchronization barriers. This paper presents a multiplexing framework that reclaims this idle capacity by co-locating multiple simulations on the same cluster. PMPI-based duty-cycle profiling quantifies the per-rank idle fraction; proportional CPU allocation then allows a second simulation to execute concurrently with minimal overhead, yielding 1.77x throughput. A Pareto sweep to N=5 concurrent simulations shows throughput scaling to 3.74x, with a knee at N=3 offering the best efficiency-cost trade-off. An analytical model with a single fitted parameter predicts these gains within +/-4%. A dynamic controller automates the full pipeline, from profiling through In-Place Pod Vertical Scaling (KEP-1287) to packing and fairness monitoring, achieving 3.25x throughput for four simulations without manual intervention or pod restarts. To our knowledge, this is the first CPU application of In-Place Pod Vertical Scaling to running MPI processes. Experiments on an AWS cluster with OpenFOAM CFD confirm that the results hold under both concentric and standard graph-based (Scotch) mesh partitioning.