Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
56 commits
Select commit Hold shift + click to select a range
55ad6ba
design: scheduler & capability model 1-pager + deliverables (preview)
May 6, 2026
471fe0d
Address Bassam's review: substrate / runtime split + node selector
May 6, 2026
e37db0a
Address Bassam's review (full): #56 claims, replica==placement, Model…
May 7, 2026
657704f
Coherence pass + revert vocab to cluster-scoped
May 7, 2026
98e5590
Collapse node+device claims, delegate to in-cluster scheduler, v1/v2 …
May 7, 2026
11cec76
Opinionated Kueue default + dual-path matching (label or DRA) + open …
May 7, 2026
1f5fa41
Make pluggable substrate concrete: BYO scheduler + backend contract
May 7, 2026
1459c11
Clarify IR == existing ModelPlacement CRD; spell out 'intermediate re…
May 7, 2026
afa1374
Refactor to broader API design framing + address Bassam's gpt-oss-20b…
May 7, 2026
036974d
Drop redundant modelplane-api/README; fold unique content into design…
May 7, 2026
2bcf6ce
Once-over: tighten adapter substrate prose, drop stale open Qs, add d…
May 7, 2026
fcd301a
Rename modelplane-api/ -> proposed-modelplane-api/; tighten diagram
May 7, 2026
8377b03
Use Bassam's whiteboard verbatim for the diagram
May 7, 2026
6d2fc9a
Reframe: federation planner vs DRA runtime allocator; instance-type m…
May 7, 2026
88eb5c5
Address Bassam's PR comments: drop requiredEngineFeatures, consumer-i…
May 7, 2026
2b120a1
Align with Bassam's GPU-hardware taxonomy + reference clusters from c…
May 7, 2026
fe9b3b2
Fine-tooth pass on examples: fix layer mismatches, stale keys, factua…
May 7, 2026
7494301
Tighten doc + add EKS-no-DRA, multi-region, L40S-4x examples
May 7, 2026
2692b5a
Reframe consumer-index discipline without comparing to a non-existent…
May 7, 2026
c6cc597
Drop accidentally committed design/.DS_Store
May 7, 2026
98f4418
Second fine-tooth pass: layering, region attrs, vendor pin, macro split
May 7, 2026
4e299bc
Add structural linter for the design examples
May 7, 2026
c7e307c
Drop Problem + Who owns what; compress Open questions to a decision t…
May 7, 2026
2829b47
Ignore design/.DS_Store
May 7, 2026
531fdf8
Ignore personal notes (*.local.md, nic-1on1.md)
May 7, 2026
92f3526
Replace CapabilityVocabulary with InferenceClass; engine catalog goes…
May 7, 2026
c08dd46
Rename ModelService → InferenceProvider; drop supportedModels
May 7, 2026
8055d0a
Make stack layering, BYO matrix, and break-glass scenarios explicit
May 7, 2026
732d723
Restructure examples by resource type; add managed/BYO/break-glass ex…
May 7, 2026
4a04c20
Final coherence pass: stale-ref + table sanity + missing rows
May 7, 2026
6086831
Drop 'Nic-owned' / 'sketch' framing on InferenceProvider
May 7, 2026
a55e1f2
Frame KServeBackend.spec.engine as a proposed extension; drop example…
May 7, 2026
21e0b85
Add KServeBackend extension XRD; drop sub-READMEs and Excalidraw
May 7, 2026
23158d9
Move modelplane-api.md under proposed-modelplane-api/
May 7, 2026
1ab78de
Sweep cleanup: stale 'deleted' framing, missing intro, LINT.md vocab ref
May 7, 2026
1ca8a00
ModelEndpoint label selectors on InferenceProvider routes; reshape mu…
May 7, 2026
e23bf1d
WIP: Nic's API sketch
negz May 7, 2026
321feab
Iterate on simplified Modelplane API sketch
negz May 8, 2026
49f6c10
Drop serving profiles from ModelDeployment
negz May 8, 2026
d668047
Rename parallelism to topology, add instances for P:D ratio
negz May 8, 2026
11fdb41
Both KAI and Kueue first-class; add placement walkthroughs + multi-te…
May 8, 2026
abf20f5
Refocus on scheduling; rename ModelPlacement→ModelReplica; defer API …
May 8, 2026
b585cfe
Drop XRDs (defer to #64); add IR + lifecycle + Quickstart sections
May 8, 2026
5f2650d
Split design doc into quickstart / advanced / scheduling / design
May 8, 2026
3a1c152
Sketch composition functions for new API; compress docs to design poi…
May 8, 2026
445e12e
KAI / Kueue scheduler integration sketch
May 8, 2026
83b984f
Modularize composition functions; add 69 unit tests
May 8, 2026
0ce26b1
gitignore Python caches + .venv-test
May 8, 2026
82c03f8
Align with K8s SIG-Scheduling vocab; document delta from main's sched…
May 8, 2026
906db4c
Drop migration framing; describe scheduler properties directly
May 8, 2026
9e35bc0
Realign composition functions to Nic's style
May 8, 2026
17d3d63
Narrow MR to managed-kai only; defer plugin system to follow-up
May 8, 2026
9e6fb04
Nest topology under workers, add workers.count for P:D ratio
negz May 8, 2026
67cd02e
Add cloud-specific provisioning to InferenceClass
negz May 9, 2026
0b21f26
Merge remote-tracking branch 'origin/pages' into dennis/scheduler-1pager
May 9, 2026
97d5d50
Update to Nic's latest API shape: workers.{count, topology}
May 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 .gitignore
Original file line number Diff line number Diff line change
@@ -1,5 +1,16 @@
.venv
.venv-test
_output
.up
result
.direnv
design/.DS_Store

# Python caches / build artifacts
__pycache__/
*.py[cod]
.pytest_cache/

# Personal notes — never commit
*.local.md
nic-1on1.md
Loading
Loading