[WIP] feat(spider-huntsman): Huntsman e2e dev branch. DO NOT MERGE!!!#370
Draft
sitaowang1998 wants to merge 80 commits into
Draft
[WIP] feat(spider-huntsman): Huntsman e2e dev branch. DO NOT MERGE!!!#370sitaowang1998 wants to merge 80 commits into
sitaowang1998 wants to merge 80 commits into
Conversation
# Conflicts: # components/spider-proto-rust/src/generated/storage.rs # components/spider-proto/storage/storage.proto # components/spider-scheduler/src/storage_client/grpc.rs
…status_to_error convention
…tus_to_error convention
…der into storage-grpc-migration
…add-missing-services
…o storage-grpc-services
… add-missing-services
# Conflicts: # components/spider-scheduler/src/error.rs # components/spider-scheduler/src/lib.rs # components/spider-scheduler/src/service.rs
# Conflicts: # components/spider-scheduler/src/storage_client/grpc.rs
…earch benchmarks. * workers: 2 -> 16. * scheduler ready_task_capacity: 1024 -> 65536, so the scheduler can hold every simultaneously-ready task across many concurrent flat jobs (N jobs x M archives) instead of backpressuring and de-fairing them. * scheduler dispatch_queue_capacity: 256 -> 32.
Runs a multi-worker CLP search over a directory of CLP archives by fanning the work across a live Spider stack (one task per archive), for benchmarking Spider on an embarrassingly-parallel search workload. * huntsman-clp-search-tasks: TDL package `clp_search` exposing `clp_search::search`, which runs `clp-s s <archive> <query>` and redirects its output to a per-task file (no task output). * huntsman-clp-search-client: discovers archives, builds a flat one-task-per-archive Spider job, submits/starts/polls it, prints the concatenated results to stdout, and reports a per-phase timing breakdown (query_processing / spider_execution / post_processing) on stderr. * huntsman-clp-search-pool-ref: a no-Spider local process-pool baseline for the same workload, used to isolate Spider's scheduling overhead. * Wire the crates into the workspace and stage `libclp_search.so` in `task build:packages`. * Add a README describing usage and internals for future reuse.
…pider into huntsman-dev-e2e
Contributor
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Pro Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
…ing metrics. * Execution manager: log per-task `scheduler_next_task_us`, `register_task_instance_us`, and `task_executor_execute_us` (structured, with job/task ids) around the scheduler fetch, the storage registration, and the task-executor round-trip. * Execution manager: propagate `RUST_LOG` to the spawned task executor so the executor and the dlopened TDL package honor the same log level. * Scheduler: log a timestamped `Dispatched a task assignment` event at dispatch, for measuring scheduling overhead (first-dispatch time vs the client's job-start time).
…le, and a clp-s-timing baseline. * Client: time `submit_job` and `start_job` separately; emit `client_job_start_epoch_us` for the scheduling-overhead calculation; shuffle archives before submission by default, with `--no-shuffle` to submit them in sorted order. * Task: time the `clp-s` subprocess and log it (structured, with job/task ids) via a package-local tracing subscriber -- a dlopened cdylib has its own tracing dispatcher, so the executor's subscriber cannot see its events. * pool-ref: shuffle archives to match the client, and print the per-archive `clp-s` execution distribution so its cost can be compared against Spider's. * Document the new flags and the per-task metrics in the README.
… for search benchmarks. * Scheduler: dispatch_queue_capacity 32 -> 64; storage_poll_timeout_ms 1000 -> 10; tick_interval_ms 10 -> 5 (react faster when short tasks drain the dispatch queue). * Storage: enable ready_queue_config with task_capacity 1048576 so many concurrent ready tasks never backpressure the ready queue.
…pider into huntsman-dev-e2e
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Checklist
breaking change.
Validation performed