Skip to content

[codex] Add GCP machine guidance for Fusion Snapshots #1273

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
llewellyn-sl wants to merge 2 commits into
base: master
Choose a base branch
from
Draft

[codex] Add GCP machine guidance for Fusion Snapshots #1273

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7310647
docs: add GCP Fusion snapshot machine guidance [EDU-1113]
llewellyn-sl Apr 3, 2026
ed4d601
Merge branch 'master' into EDU-1113-docs-draft-v2
justinegeffen Apr 9, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 11 additions & 0 deletions fusion_docs/guide/snapshots/gcp.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,17 @@ Fusion Snapshots work with sensible defaults (5 automatic retry attempts). For c

[Incremental snapshots](./index.md#incremental-snapshots) are enabled by default on x86_64 instances and capture only changed memory pages between checkpoints. This is particularly beneficial for Google Batch's shorter reclamation window. Use x86_64 instances to enable incremental snapshots.

## Machine type guidance

Fusion Snapshots on Google Batch work best when the underlying compute environment uses machine types that provide local SSD support and enough memory bandwidth to complete checkpoints within the preemption window.

- If you don't specify a machine type, Seqera Platform selects a VM from Google Cloud families that support local SSDs.
- Any machine type you specify for Fusion Snapshots must support local SSDs.
- For production workloads, start with an `n2-highmem-16-lssd` VM or larger, then validate checkpoint duration with your workload profile.
- If your workload has larger memory footprints, increase the machine size conservatively and re-test snapshot and restore times before widening usage.

See [Google Cloud Batch compute environment configuration](../../platform-cloud/docs/compute-envs/google-cloud-batch.md#use-fusion-v2) for the underlying Fusion v2 compute recommendations that also apply to Fusion Snapshots on Google Batch.

## Resource limits

A single job can request more resources than are available on a single instance. To prevent this, set resource limits using the `process.resourceLimits` directive in your Nextflow configuration. See [Resource limits](./configuration.md#resource-limits) for more information.
Expand Down
Loading