feat(operator): add experimental grove.forceScalingGroup for single-node components by joeltg · Pull Request #11772 · ai-dynamo/dynamo

joeltg · 2026-07-16T15:24:22Z

Overview:

Adds an opt-in, per-component experimental.grove.forceScalingGroup boolean (v1beta1) that lets a single-node component render as a Grove PodCliqueScalingGroup instead of a standalone PodClique, so each replica becomes an independently gang-scheduled unit and horizontal scaling changes the scaling-group replica count.

API shape follows the maintainer proposal in #11774: false or omitted means "use automatic selection" (multi-node and inter-pod GMS components use a PCSG, ordinary single-node components use a PodClique), not "force PodClique".

Details:

Today the PCSG decision is hardcoded: only multi-node and inter-pod GMS components render as a PCSG; single-node components pool all replicas as pods of one standalone PodClique. With this change:

components:
- componentName: worker
  replicas: 4
  minAvailable: 1
  experimental:
    grove:
      forceScalingGroup: true

renders one PodCliqueScalingGroupConfig (replicas: 4, minAvailable: 1) containing a single one-pod clique. The scheduler sees one base PodGang (PCSG replicas [0, minAvailable) plus standalone cliques) and an independent scaled PodGang per remaining replica, each with a single one-pod PodGroup. Scaling the component patches the PCSG replicas, adding/removing whole PodGangs.

API: experimental.grove.forceScalingGroup field; UsesPCSG() (= isMultinode || isInterPodGMS || forceScalingGroup) centralizes the previously duplicated PCSG decision (PCS rendering, Grove readiness aggregation, restart progress tracking, reconcileGroveScaling, resource-name-length validation).
Rendering: single-node forced components emit one RoleMain clique with replicas: 1; the existing PCSG config path carries component.replicas / minAvailable unchanged. forceScalingGroup: true on a multi-node component is a redundant no-op (automatic selection already yields a PCSG).
Validation: Grove pathway only (same rule as minAvailable); the effective opt-in is immutable after creation (same pattern as the inter-pod GMS layout toggle since it re-shapes rendered resource names); explicit false is equivalent to omitted and can be added/removed freely.
Conversion: the grove block has no v1alpha1 representation; it is preserved across spoke round-trips, sparsely when set alongside alpha-representable experimental fields.
Generated: CRDs and deepcopy regenerated.

Defaults are fully backward compatible: omitting the field keeps today's rendering exactly.

Where should the reviewer start?

deploy/operator/api/v1beta1/common.go — the field and its semantics
deploy/operator/internal/dynamo/graph.go — expandSingleNodeScalingGroupRoles + UsesPCSG() call sites
deploy/operator/internal/webhook/validation/shared_v1beta1.go — create/update rules
deploy/operator/api/v1alpha1/shared_spec_conversion.go — hub-only preservation

Related Issues

🔗 This PR is linked to an issue:

Relates to DEP (light): Experimental replicaGrouping for single-node Grove components #11774

🤖 Generated with Claude Code

Summary by CodeRabbit

New Features
- Added the experimental Grove forceScalingGroup option for rendering single-node components with scaling-group support.
- Added automatic selection of scaling-group rendering based on component configuration.
Validation
- Restricted the option to supported Grove-backed components.
- Made the setting immutable after creation.
Documentation
- Updated CRD schemas and API reference documentation with configuration details.
Bug Fixes
- Preserved Grove configuration during API version conversions.

copy-pr-bot · 2026-07-16T15:24:26Z

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

github-actions · 2026-07-16T15:24:38Z

🌿 Fern Docs Preview: https://nvidia-preview-019fd3bc-8048-74f2-bef1-cf170df57d3d.docs.buildwithfern.com/dynamo/dev

devin-ai-integration

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no bugs or issues to report.

coderabbitai · 2026-07-18T16:11:47Z

Walkthrough

The PR adds spec.experimental.grove.forceScalingGroup to the v1beta1 API. It preserves the field during conversion, validates Grove usage and immutability, and routes Grove rendering and reconciliation through UsesPCSG().

Changes

Grove force scaling group

Layer / File(s)	Summary
API shape and conversion preservation `deploy/operator/api/v1beta1/...`, `deploy/operator/api/v1alpha1/...`, `deploy/operator/config/crd/bases/...`, `docs/fern/...`	Adds `GroveSpec`, CRD schemas, deepcopy methods, API documentation, conversion preservation, and round-trip coverage.
Validation and update immutability `deploy/operator/internal/webhook/validation/...`	Validates Grove pathway usage and rejects effective `forceScalingGroup` changes after creation. Envtest cases cover create and update behavior.
PCSG selection and Grove rendering `deploy/operator/api/v1beta1/dynamocomponentdeployment_types.go`, `deploy/operator/internal/dynamo/...`, `deploy/operator/internal/webhook/validation/dynamographdeployment_helpers.go`	Adds `UsesPCSG()` and uses it for role expansion, naming, readiness-resource selection, and Grove PodCliqueSet generation. Forced single-node components use one engine pod and horizontal PCSG replicas.
Reconciliation and readiness routing `deploy/operator/internal/controller/...`	Uses `UsesPCSG()` to select Grove readiness checks and scaling targets.

Estimated code review effort: 4 (Complex) | ~45 minutes

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely identifies the addition of experimental grove.forceScalingGroup for single-node components.
Description check	✅ Passed	The description completes all required sections and explains the API, rendering, validation, conversion, generated changes, and review starting points.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

Warning

There were issues while running some tools. Please review the errors and either fix the tool's configuration or disable the tool if it's a critical failure.

🔧 golangci-lint (2.12.2)

level=error msg="[linters_context] typechecking error: pattern ./...: directory prefix . does not contain main module or its selected dependencies"

_{Comment @coderabbitai help to get the list of available commands.}

coderabbitai

🧹 Nitpick comments (1)

deploy/operator/api/v1alpha1/dynamocomponentdeployment_conversion_test.go (1)
626-668: 🗄️ Data Integrity & Integration | 🔵 Trivial | ⚡ Quick win

Add graph-level conversion round-trip coverage.

This test only exercises DynamoComponentDeployment, while DynamoGraphDeploymentSpec.components.experimental.grove.forceScalingGroup is also a conversion contract. Add a graph round-trip case to protect that path independently.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@deploy/operator/api/v1alpha1/dynamocomponentdeployment_conversion_test.go`
around lines 626 - 668, Extend TestDCD_RoundTrip_ExperimentalGrove with an
independent DynamoGraphDeploymentSpec round-trip case covering
components.experimental.grove.forceScalingGroup. Construct a graph containing
the grove-only experimental field, run it through the graph conversion
round-trip helper, and assert the original and converted graph specs are
equivalent.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@deploy/operator/api/v1alpha1/dynamocomponentdeployment_conversion_test.go`:
- Around line 626-668: Extend TestDCD_RoundTrip_ExperimentalGrove with an
independent DynamoGraphDeploymentSpec round-trip case covering
components.experimental.grove.forceScalingGroup. Construct a graph containing
the grove-only experimental field, run it through the graph conversion
round-trip helper, and assert the original and converted graph specs are
equivalent.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 0fad3eeb-6acc-4f01-a596-4b928b567926

📥 Commits

Reviewing files that changed from the base of the PR and between dad08ea and 8acaa5a.

📒 Files selected for processing (16)

deploy/operator/api/v1alpha1/conversion_field_coverage_test.go
deploy/operator/api/v1alpha1/dynamocomponentdeployment_conversion_test.go
deploy/operator/api/v1alpha1/shared_spec_conversion.go
deploy/operator/api/v1beta1/common.go
deploy/operator/api/v1beta1/dynamocomponentdeployment_types.go
deploy/operator/api/v1beta1/zz_generated.deepcopy.go
deploy/operator/config/crd/bases/nvidia.com_dynamocomponentdeployments.yaml
deploy/operator/config/crd/bases/nvidia.com_dynamographdeployments.yaml
deploy/operator/internal/controller/dynamographdeployment_controller.go
deploy/operator/internal/dynamo/graph.go
deploy/operator/internal/dynamo/graph_test.go
deploy/operator/internal/dynamo/grove.go
deploy/operator/internal/webhook/validation/dynamographdeployment_helpers.go
deploy/operator/internal/webhook/validation/dynamographdeployment_test.go
deploy/operator/internal/webhook/validation/shared_helpers.go
deploy/operator/internal/webhook/validation/shared_v1beta1.go

…ode components Add components[].experimental.grove.forceScalingGroup (bool) to the v1beta1 shared component spec. False or omitted means automatic selection: multi-node and inter-pod GMS components render as a Grove PodCliqueScalingGroup, other single-node components pool their replicas as pods of one standalone PodClique — exactly today's behavior. Setting forceScalingGroup: true opts a single-node component into the PCSG layout: the scaling-group replica count carries the horizontal scale, with a single one-pod PodClique per replica, so each replica is an independently gang-scheduled unit (its own PodGang beyond minAvailable) and scaling stamps out or removes whole PodGangs. - Centralize the previously duplicated "multinode || inter-pod GMS" PCSG decision as DynamoComponentDeploymentSharedSpec.UsesPCSG() (= isMultinode || isInterPodGMS || forceScalingGroup) and use it in PCS rendering, Grove readiness aggregation, restart progress tracking, reconcileGroveScaling, and resource-name-length validation. - Validation: Grove pathway only (same rule as minAvailable); the effective opt-in is immutable after creation (same pattern as the inter-pod GMS layout toggle); explicit false is equivalent to omitted and freely add/removable. - Conversion: the grove block has no v1alpha1 representation; preserve it across spoke round-trips, sparsely when it sits alongside alpha-representable experimental fields. API shape per review discussion in ai-dynamo#11774. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Signed-off-by: joeltg <joel@reflection.ai>

julienmancuso

Two P2 findings from the operator API and validation review.

joeltg · 2026-07-30T12:24:27Z

@julienmancuso any other feedback here? would love to get this in, would give us a lot more confidence in our ability to integrate with our scheduler :)

athreesh · 2026-08-04T00:09:50Z

@joeltg reading thru PR, given offline context

julienmancuso · 2026-08-04T15:40:30Z

@joeltg , sorry for the delay, please rebase and i'll review / approve with high priority

…eplica-grouping Signed-off-by: joeltg <joel@reflection.ai> # Conflicts: # deploy/operator/internal/controller/dynamographdeployment_controller.go # deploy/operator/internal/dynamo/grove.go

julienmancuso · 2026-08-04T17:29:55Z

/ok to test e064cf5

julienmancuso · 2026-08-04T17:30:07Z

@coderabbitai full review

coderabbitai · 2026-08-04T17:30:13Z

✅ Action performed

Full review finished.

coderabbitai

🧹 Nitpick comments (4)

deploy/operator/internal/webhook/validation/shared_helpers.go (1)
185-190: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low value

Add story comments to the new multi-line Go blocks.

deploy/operator/internal/webhook/validation/shared_helpers.go#L185-L190: Add a comment that explains the nil ExperimentalSpec behavior.

deploy/operator/internal/webhook/validation/shared_v1beta1.go#L232-L238: Add a comment that explains Grove create validation.

deploy/operator/internal/webhook/validation/shared_v1beta1.go#L315-L320: Add a comment that explains the Grove-pathway rejection.

deploy/operator/internal/webhook/validation/shared_v1beta1.go#L432-L438: Add a comment that explains removal handling.

deploy/operator/internal/webhook/validation/shared_v1beta1.go#L515-L528: Add a comment that explains Grove update validation.

deploy/operator/internal/webhook/validation/shared_v1beta1.go#L541-L549: Add a comment that explains effective opt-in immutability.

As per coding guidelines, “In Go code, put a one-line story comment above every multi-line block of logically connected code.”
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@deploy/operator/internal/webhook/validation/shared_helpers.go` around lines
185 - 190, Add one-line story comments above each specified multi-line block: in
deploy/operator/internal/webhook/validation/shared_helpers.go lines 185-190,
explain the nil ExperimentalSpec behavior; in
deploy/operator/internal/webhook/validation/shared_v1beta1.go lines 232-238,
315-320, 432-438, 515-528, and 541-549, respectively explain Grove create
validation, Grove-pathway rejection, removal handling, Grove update validation,
and effective opt-in immutability. Keep the comments concise and describe the
block’s behavior without changing its logic.
Source: Coding guidelines
deploy/operator/internal/dynamo/graph_test.go (1)
5085-5100: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Add t.Log headings for the test steps.

The new tests use comments but do not log their setup, render, and assertion phases.

deploy/operator/internal/dynamo/graph_test.go#L5085-L5100: Add t.Log headings before component setup, role expansion, and result comparison.

deploy/operator/internal/dynamo/graph_test.go#L8793-L8847: Add t.Log headings before deployment setup, beta-only configuration, rendering, and PCSG assertions.

As per coding guidelines, “In Go tests, use t.Log to explain the test's story, with one heading before each block implementing a test step.”
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@deploy/operator/internal/dynamo/graph_test.go` around lines 5085 - 5100, Add
t.Log headings for each test step in
deploy/operator/internal/dynamo/graph_test.go:5085-5100, before component setup,
role expansion, and result comparison in
TestExpandRolesForComponent_SingleNodeForceScalingGroup; and in
deploy/operator/internal/dynamo/graph_test.go:8793-8847, before deployment
setup, beta-only configuration, rendering, and PCSG assertions. Use one heading
per block without changing test behavior.
Source: Coding guidelines
deploy/operator/api/v1alpha1/dynamocomponentdeployment_conversion_test.go (1)
626-628: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Use t.Log for the test story.

This new test uses a paragraph comment to describe its steps. The Go test guidelines require t.Log headings before test-step blocks. Move the story into the subtest before fixture construction and round-trip validation.
Proposed change
-// The grove block has no v1alpha1 representation and must survive the spoke
-// round-trip both alone (whole hub-only block preserved) and alongside
-// alpha-representable fields (sparse preservation merged back).
 func TestDCD_RoundTrip_ExperimentalGrove(t *testing.T) {
...
 		t.Run(tt.name, func(t *testing.T) {
+			t.Log("Round-trip the Grove configuration through v1alpha1 conversion")
 			src := &v1beta1.DynamoComponentDeployment{
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@deploy/operator/api/v1alpha1/dynamocomponentdeployment_conversion_test.go`
around lines 626 - 628, Update the affected conversion test subtest to replace
the paragraph comment describing the grove round-trip scenarios with a t.Log
heading before fixture construction and round-trip validation, preserving the
existing test behavior and story content.
Source: Coding guidelines
deploy/operator/api/v1alpha1/conversion_field_coverage_test.go (1)
121-121: 🗄️ Data Integrity & Integration | 🔵 Trivial | ⚡ Quick win

Add graph-deployment round-trip coverage.

The field set now acknowledges DynamoGraphDeploymentSpec.components.experimental.grove.forceScalingGroup, but the supplied round-trip test covers only DynamoComponentDeployment. Add a graph-deployment round-trip case, or verify an existing test exercises this conversion path. Otherwise, a graph conversion regression could drop the preserved field while this acknowledgement test still passes.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@deploy/operator/api/v1alpha1/conversion_field_coverage_test.go` at line 121,
Add round-trip coverage for DynamoGraphDeploymentSpec conversion, specifically
exercising components.experimental.grove.forceScalingGroup through both
conversion directions and verifying it is preserved. Update the existing
round-trip test suite in conversion_field_coverage_test.go or add a dedicated
graph-deployment case alongside the DynamoComponentDeployment coverage.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@deploy/operator/api/v1alpha1/conversion_field_coverage_test.go`:
- Line 121: Add round-trip coverage for DynamoGraphDeploymentSpec conversion,
specifically exercising components.experimental.grove.forceScalingGroup through
both conversion directions and verifying it is preserved. Update the existing
round-trip test suite in conversion_field_coverage_test.go or add a dedicated
graph-deployment case alongside the DynamoComponentDeployment coverage.

In `@deploy/operator/api/v1alpha1/dynamocomponentdeployment_conversion_test.go`:
- Around line 626-628: Update the affected conversion test subtest to replace
the paragraph comment describing the grove round-trip scenarios with a t.Log
heading before fixture construction and round-trip validation, preserving the
existing test behavior and story content.

In `@deploy/operator/internal/dynamo/graph_test.go`:
- Around line 5085-5100: Add t.Log headings for each test step in
deploy/operator/internal/dynamo/graph_test.go:5085-5100, before component setup,
role expansion, and result comparison in
TestExpandRolesForComponent_SingleNodeForceScalingGroup; and in
deploy/operator/internal/dynamo/graph_test.go:8793-8847, before deployment
setup, beta-only configuration, rendering, and PCSG assertions. Use one heading
per block without changing test behavior.

In `@deploy/operator/internal/webhook/validation/shared_helpers.go`:
- Around line 185-190: Add one-line story comments above each specified
multi-line block: in
deploy/operator/internal/webhook/validation/shared_helpers.go lines 185-190,
explain the nil ExperimentalSpec behavior; in
deploy/operator/internal/webhook/validation/shared_v1beta1.go lines 232-238,
315-320, 432-438, 515-528, and 541-549, respectively explain Grove create
validation, Grove-pathway rejection, removal handling, Grove update validation,
and effective opt-in immutability. Keep the comments concise and describe the
block’s behavior without changing its logic.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: a828284e-ca37-4156-b082-f4da91796c1e

📥 Commits

Reviewing files that changed from the base of the PR and between 25090d2 and e064cf5.

📒 Files selected for processing (19)

deploy/operator/api/v1alpha1/conversion_field_coverage_test.go
deploy/operator/api/v1alpha1/dynamocomponentdeployment_conversion_test.go
deploy/operator/api/v1alpha1/shared_spec_conversion.go
deploy/operator/api/v1beta1/common.go
deploy/operator/api/v1beta1/dynamocomponentdeployment_types.go
deploy/operator/api/v1beta1/zz_generated.deepcopy.go
deploy/operator/config/crd/bases/nvidia.com_dynamocomponentdeployments.yaml
deploy/operator/config/crd/bases/nvidia.com_dynamographdeployments.yaml
deploy/operator/internal/controller/dgd_grove_restart_progress_resolver.go
deploy/operator/internal/controller/dgd_grove_scaler.go
deploy/operator/internal/dynamo/graph.go
deploy/operator/internal/dynamo/graph_test.go
deploy/operator/internal/dynamo/grove.go
deploy/operator/internal/webhook/validation/dynamographdeployment_helpers.go
deploy/operator/internal/webhook/validation/dynamographdeployment_validation_envtest_test.go
deploy/operator/internal/webhook/validation/shared_helpers.go
deploy/operator/internal/webhook/validation/shared_test.go
deploy/operator/internal/webhook/validation/shared_v1beta1.go
docs/fern/pages/reference/kubernetes-api/additional-resources/api-reference-k8s.md

julienmancuso

Two additional P2 findings from the Grove scheduling/API contract review.

…eplica-grouping Signed-off-by: joeltg <joel@reflection.ai> # Conflicts: # deploy/operator/internal/webhook/validation/shared_test.go # deploy/operator/internal/webhook/validation/shared_v1beta1.go

…apping accurately Address review feedback on the published API contract: - minAvailable: describe the Grove mapping in terms of scaling-group membership (multi-node, inter-pod GMS, or grove.forceScalingGroup; see UsesPCSG) instead of the stale single-node-vs-multi-node split. - forceScalingGroup: state the factual PodGang mapping — the first minAvailable replicas join the deployment's base PodGang together with its other base workloads — instead of promising fully independent replicas, which Grove's base-gang semantics cannot provide. Regenerate CRDs and the API reference from the updated comments. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Signed-off-by: joeltg <joel@reflection.ai>

julienmancuso · 2026-08-04T23:28:38Z

/ok to test a46cbda

joeltg · 2026-08-05T03:28:13Z

^ CI just needs a rerun for the failed workflow I think! thanks so much @julienmancuso

dynamo-ops · 2026-08-05T13:24:48Z

/ok to test e6d6b52

julienmancuso · 2026-08-05T13:24:48Z

/ok to test e6d6b52

harryskim

LGTM

julienmancuso · 2026-08-05T20:58:03Z

/ok to test 69faafd

dynamo-ops · 2026-08-05T20:59:48Z

/ok to test 69faafd

dyn-3691-extract-shared-target-pid-cuda-customstorage-operation-layer * 'main' of https://github.com/ai-dynamo/dynamo: (65 commits) fix(frontend): emit SGLang stream role once (#12741) docs(fern): promote v1.3.1 to current release (#12752) fix(docs): remove duplicate unscoped community-rail CSS rules (#12615) feat(operator): migrate CRD storage to v1beta1 (#11904) fix: synchronize self-benchmark capacity across DP ranks (#12021) chore(deps): bump dynamo-tokenizers to 1.8.0 (#12707) fix(frontend): preserve split UTF-8 characters (#12688) docs: align Kubernetes build selector with CLI (#12729) fix(frontend): preserve completion backend error status (#12706) fix(operator): replace snapshot pods after GMS restart (#11286) refactor(media): rename installer module, drop --packages per review fix(media): harden installer against three pre-redesign review findings fix(media): verify installs in a fresh interpreter; teach --pip-args= form test(serve): install test-time decoders at the validated bounds feat(media): explicit installer for additional media decoders docs(spica): correct kv_load_ratio support guidance (#12714) feat(operator): add experimental grove.forceScalingGroup for single-node components (#11772) fix(vllm): declare entry-stage engine_input_source in GLM-Image NIXL config (#12709) chore: bump trtllm to v1.3.0rc23 (#12532) perf: remove trtllm postprocessing workers from the args as post processing workers are not effective in dynamo (#12592) ... Signed-off-by: Hannah Zhang <hannahz@nvidia.com>

joeltg temporarily deployed to external_collaborator July 16, 2026 15:24 — with GitHub Actions Inactive

pull-request-size Bot added the size/L label Jul 16, 2026

github-actions Bot added external-contribution Pull request is from an external contributor deployment::k8s Relates to dynamo deployment in kubernetes feat labels Jul 16, 2026

joeltg force-pushed the feat/experimental-replica-grouping branch from 1ef49bd to 5060ac2 Compare July 16, 2026 15:35

joeltg temporarily deployed to external_collaborator July 16, 2026 15:36 — with GitHub Actions Inactive

This comment has been minimized.

Sign in to view

joeltg mentioned this pull request Jul 16, 2026

DEP (light): Experimental replicaGrouping for single-node Grove components #11774

Closed

joeltg force-pushed the feat/experimental-replica-grouping branch from 5060ac2 to 8acaa5a Compare July 18, 2026 15:21

joeltg temporarily deployed to external_collaborator July 18, 2026 15:21 — with GitHub Actions Inactive

joeltg changed the title ~~feat(operator): add experimental replicaGrouping for single-node Grove components~~ feat(operator): add experimental grove.forceScalingGroup for single-node components Jul 18, 2026

joeltg marked this pull request as ready for review July 18, 2026 16:04

joeltg requested a review from a team as a code owner July 18, 2026 16:04

devin-ai-integration Bot reviewed Jul 18, 2026

View reviewed changes

coderabbitai Bot reviewed Jul 18, 2026

View reviewed changes

joeltg force-pushed the feat/experimental-replica-grouping branch from 8acaa5a to 62654f0 Compare July 18, 2026 16:15

joeltg temporarily deployed to external_collaborator July 18, 2026 16:15 — with GitHub Actions Inactive

joeltg force-pushed the feat/experimental-replica-grouping branch from 62654f0 to 40464ac Compare July 18, 2026 16:28

joeltg requested a review from a team as a code owner July 18, 2026 16:28

joeltg temporarily deployed to external_collaborator July 18, 2026 16:28 — with GitHub Actions Inactive

github-actions Bot added the documentation Improvements or additions to documentation label Jul 18, 2026

Merge branch 'main' into feat/experimental-replica-grouping

e04a0bf

joeltg temporarily deployed to external_collaborator July 20, 2026 13:20 — with GitHub Actions Inactive

julienmancuso reviewed Jul 21, 2026

View reviewed changes

Comment thread deploy/operator/internal/webhook/validation/shared_v1beta1.go Outdated

Comment thread deploy/operator/api/v1beta1/common.go Outdated

Merge remote-tracking branch 'upstream/main' into feat/experimental-r…

e064cf5

…eplica-grouping Signed-off-by: joeltg <joel@reflection.ai> # Conflicts: # deploy/operator/internal/controller/dynamographdeployment_controller.go # deploy/operator/internal/dynamo/grove.go

joeltg temporarily deployed to external_collaborator August 4, 2026 17:26 — with GitHub Actions Inactive

coderabbitai Bot reviewed Aug 4, 2026

View reviewed changes

julienmancuso approved these changes Aug 4, 2026

View reviewed changes

julienmancuso reviewed Aug 4, 2026

View reviewed changes

Comment thread deploy/operator/api/v1beta1/common.go Outdated

Comment thread deploy/operator/api/v1beta1/dynamocomponentdeployment_types.go

joeltg and others added 2 commits August 4, 2026 18:18

Merge remote-tracking branch 'upstream/main' into feat/experimental-r…

7b7a1a4

…eplica-grouping Signed-off-by: joeltg <joel@reflection.ai> # Conflicts: # deploy/operator/internal/webhook/validation/shared_test.go # deploy/operator/internal/webhook/validation/shared_v1beta1.go

joeltg temporarily deployed to external_collaborator August 4, 2026 18:20 — with GitHub Actions Inactive

julienmancuso approved these changes Aug 4, 2026

View reviewed changes

julienmancuso enabled auto-merge (squash) August 4, 2026 23:31

Merge branch 'main' into feat/experimental-replica-grouping

e6d6b52

julienmancuso temporarily deployed to external_collaborator August 5, 2026 13:24 — with GitHub Actions Inactive

harryskim approved these changes Aug 5, 2026

View reviewed changes

Merge branch 'main' into feat/experimental-replica-grouping

69faafd

julienmancuso temporarily deployed to external_collaborator August 5, 2026 20:57 — with GitHub Actions Inactive

julienmancuso merged commit 5c59b3e into ai-dynamo:main Aug 5, 2026
96 checks passed

dagil-nvidia mentioned this pull request Aug 10, 2026

docs: backport main docs tree and docs CI to release/1.4.0 #12950

Merged

Uh oh!

Conversation

joeltg commented Jul 16, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Overview:

Details:

Where should the reviewer start?

Related Issues

Summary by CodeRabbit

Uh oh!

copy-pr-bot Bot commented Jul 16, 2026

Uh oh!

github-actions Bot commented Jul 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment has been minimized.

devin-ai-integration Bot left a comment

Choose a reason for hiding this comment

✅ Devin Review: No Issues Found

Uh oh!

coderabbitai Bot commented Jul 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

❌ Failed checks (1 warning)

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

julienmancuso left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

joeltg commented Jul 30, 2026

Uh oh!

athreesh commented Aug 4, 2026

Uh oh!

julienmancuso commented Aug 4, 2026

Uh oh!

julienmancuso commented Aug 4, 2026

Uh oh!

julienmancuso commented Aug 4, 2026

Uh oh!

coderabbitai Bot commented Aug 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

julienmancuso left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

julienmancuso commented Aug 4, 2026

Uh oh!

joeltg commented Aug 5, 2026

Uh oh!

dynamo-ops commented Aug 5, 2026

Uh oh!

julienmancuso commented Aug 5, 2026

Uh oh!

harryskim left a comment

Choose a reason for hiding this comment

Uh oh!

julienmancuso commented Aug 5, 2026

Uh oh!

dynamo-ops commented Aug 5, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

joeltg commented Jul 16, 2026 •

edited by coderabbitai Bot

Loading

github-actions Bot commented Jul 16, 2026 •

edited

Loading

coderabbitai Bot commented Jul 18, 2026 •

edited

Loading

coderabbitai Bot commented Aug 4, 2026 •

edited

Loading