Skip to content

Load Balancing Distributed GPU Computations #5845

Description

@jackdfranklin

Benchmarking distributed GPU computations with nsight systems indicates that the bottleneck to improving strong scaling for simulations on immersed grids is now load balancing.

This arises from the use of the active cells map launching, which only performs computations on cells in the grid that are not immersed. For most bathymetrys (e.g. earth bathymetry) the naive partitioning of the grid into equal sized cells results in sub-grids containing very different amounts of active cells. This results in poor load balancing, causing the GPUs to become desynchronised and reducing utilisation.

I am opening this issue to track the outcome of different partitioning algorithms and their performance characteristics. Eventually this will result in a PR to add automatically balanced grid partitioning.

Metadata

Metadata

Assignees

Labels

Launch Distributed BuildkiteSelect this label to run the Distributed Buildkite testsdistributed 🕸️Our plan for total cluster dominationfeature 🌟Something new and shinyperformance 🏍️So we can get the wrong answer even faster

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions