Distribute only each stage's GGUF layer slice over HTTP, add sampler and weight-load health telemetry, and harden node-image build, orchestrator provisioning, and the VastAI lease/search path.
- gguf_shard (new): StageShardPlan and plan_stage_shard parse the GGUF directory and compute coalesced per-stage tensor byte ranges; materialize_stage_shard_http fetches only those ranges (plus the header) to build a stage-local GGUF, with planned_fetch_bytes accounting.
- orchestrator_app: build a BTreeMap<u32, StageShardPlan> from the run plan for HuggingFace sources, thread stage_shard_plan through StageProvisionWire and weight-load, emit stage_shard_plan summaries, and add liveness phases (prefetching/fetching_stage_shard, cache_ready, stage_shard_ready).
- worker_node: add a stage-shard-fetcher subcommand and materialize_stage_shard_with_process that spawns the fetcher, streams its stdout/stderr as stage_shard_fetch events (StageShardCacheReady/StageShardReady), caches under MVP_MODEL_CACHE_DIR, and feeds the local shard path into load_weights.
- worker_node: add NODE_SAMPLER_CHANNEL and SamplerHealth telemetry (gpu/cpu/net samplers emit started/waiting/ready/failed) plus structured helper stdout/stderr streaming (wait_for_helper_event/drain_worker_stderr).
- node_image: expand node-image build/push handling for the deploy path.
- tools/vastai: extend lease, search, and types and drop unused pricing code.
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
Complete an 8-stage pipeline-parallel run over VastAI by provisioning stages high-to-low, adding per-stage/per-step metrics, host anti-colocation, and provider state-timeout guardrails.
- orchestrator_app: select the next weight-load stage by max index (provision stages high-to-low for parallel spread), add a throttled "loaded N of M; waiting on stage X" stage_provision_wait headline, and surface min_compute_cap/state_timeout_secs in the config dump.
- orchestrator_app: enrich pipeline_token_in/out and tokenizer_decode events with token_count/token_ids/generated_index.
- worker_node: add timing metrics across the data path (helper_execute_ms, egress_ring_read_ms, send_ms, ingress_ring_write_ms, object_load_ms), refactor take_complete_ingress_record into IngressRecordBytes (object_id/sequence/extent/flags), and emit a new object_loaded event.
- vastai_provisioning: track leased host_ids and blacklist already-leased hosts in later ProvisionRequests so stages don't co-locate, and tag SSH-bootstrap retry logs with the attempt number.
- tools/vastai: add min_compute_cap (PP_MIN_COMPUTE_CAP) filter/search query and a LifecyclePolicy state_timeout (PP_STATE_TIMEOUT_SECS) that fails instances stuck in a non-running status instead of polling forever.
- xtask: raise the check timeout to 1800s/30s grace, drop --skip-rebuild for VastAI, aggregate per-stage StepExecuted metrics, add a vastai summary section, and write failure artifacts on abort.
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
Land the first working two-stage pipeline-parallel run over VastAI, wiring a real inter-stage data path with observability, a max-price offer cap, and remote-image reuse.
- orchestrator_app: raise the VastAI pipeline-stage cap from 1 to 2 and let VastAI pipeline planning resolve the HuggingFace GGUF from the default local cached-model metadata path instead of requiring host mounts; add --vastai-max-dph-total (CLI/env/TOML) config.
- vastai_provisioning: make complete_bootstrap a no-op so the SSH bootstrap log tail stays alive past runtime-ready until node stop, preserving post-ready worker logs; add a test asserting the tail is only stopped on NodeStop.
- worker_node: emit data-path NodeEvents across the pipeline (iroh_edge_stream_arrived/bytes_read/bytes_sent, egress_ring_read, ingress_ring_write) with edge/byte metadata.
- tools/vastai: add max_dph_total (PP_MAX_DPH_TOTAL) to SelectionPolicy, the reachable-offer filter, and the search query, and improve the empty-pool error message.
- xtask: pass --skip-rebuild for the VastAI scenario and gate it on a new require_vastai_data_path_facts plus GPU facts (ring install, activation object load/step, interstage handoff, iroh edge read/sent).
- mvp_chat: add ChatModelConfig (model id/gguf/tokenizer/max-context) forwarded to the orchestrator; for VastAI + skip-rebuild, emit skip events and reuse the remote node image without a local build.
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
Move iroh_driver and the relay binary out of distribution into a dedicated
crates/iroh-driver (lib re-exports IrohDriver; relay bin renamed). Remove the node crate
and the single-gpu-inference example; drop the docker/datastream demo. Slim
pipeline-parallel vastai.
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
Remove co-dependencies for different modules found in `crates` and migrate the development history to a new repository. The docs were stale, and largely not getting used, so simply deleted for now. When code stabilizes more, they will become useful again.
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
Extract the timer wheel and watch registry from the worker into a reusable std-extension crate, generalize the worker around a RuntimeExtension factory, and decompose the monolithic runtime test file into focused suites.
- crates/std: extract TimerWheel (deterministic tick-counted one-shot/interval timers) and WatchRegistry (target-to-watcher death-notification index) out of src/worker.rs into reusable modules
- crates/std: add Ctx extension traits (CtxMonitoring, CtxNaming, CtxWatching, CtxTimers) and Runtime extension traits (RuntimeNaming, RuntimeWatching, RuntimeGroups) wiring monitor/name/watch/timer/group support
- src/worker.rs: replace the hard-coded timer/watch fields with a generic RuntimeExtension factory and add route_to_pool_or_remote for message routing (local pool, then cross-worker address map, then external inboxes)
- tests: split the 4622-line tests/runtime_api.rs into focused suites (actor_lifecycle, message_delivery, runtime_stress, std_extension) plus a shared tests/common/mod.rs harness, and drop watch_api.rs
- benches/fuzz: add runtime_benchmarks and adjust the runtime fuzz target
- tools/docs: add fn_complexity.py and loc_analysis.py analysis scripts and refresh the runtime and worker-thread docs
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
Split the monolithic crate into a Cargo workspace with the Python and Wasm bindings as separate member crates.
- Cargo.toml: declare a `[workspace]` with members `.`/`crates/swactor-python`/`crates/swactor-wasm`, remove the `python` feature and pyo3 dependency, and change root crate-type from `["cdylib","rlib"]` to `["rlib"]`
- crates/swactor-python: new cdylib crate re-exporting the PyO3 bindings (Runtime/RuntimeConfig/RuntimeHandle/Inbox/Ctx/ActorAddress/RuntimeStats), depending on `swactor` + pyo3; pyproject.toml and uv.lock relocated here from the root
- crates/swactor-wasm: new cdylib crate moved from top-level `wasm/`, depending on `swactor` with `no_random` features
- src/actor.rs: widen `Actor::new`, `AnyActor`, `ContextInner`, and `Ctx::raw_inner` to `pub` so the separate binding crates can drive the runtime
- src/lib.rs: delete the in-tree `python` module and `#[pymodule]`, and gate the `no_random` RNG behind `all(feature = "no_random", not(feature = "getrandom"))`
- tools/: relocate package.json/package-lock.json; drop the now-duplicate `wasm/Cargo.lock`
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
Asking the agent to refactor to reduce spectral complexity, it worked. Trivial change, but this did reduce code complexity.
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
Generates a report on the connectivity properties of the code interdependency DAG.
Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
