-
Notifications
You must be signed in to change notification settings - Fork 1
FCFS load balancer (central queue) with Erlang-C validation, proper Wq measurement, and metrics plumbing #29
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.
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
Merged
Conversation
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
* setting up the generator * poisson-poisson requests sampling * Delete docker_fs/.env.dev * Delete docker_fs/.env.test * changes * refactor for event generator * added gaussian truncated generator and unit tests for helpers * imprved documentantion and simulation time handling * simulation start with the vent generator * small refactor + tests for the generation of events * minor changes * Update src/app/schemas/simulation_input.py Co-authored-by: Copilot <[email protected]> * Update tests/unit/sampler/test_sampler_helper.py Co-authored-by: Copilot <[email protected]> * Update scripts/quality-check.sh Co-authored-by: Copilot <[email protected]> * Update tests/integration/db_initialization/test_init_models.py Co-authored-by: Copilot <[email protected]> * minor changes after PR review --------- Co-authored-by: Copilot <[email protected]>
* moving file to correct folder * Documentation added * Update requests_generator.md * test Added, introduced constants for the sampling window * Update tests/unit/sampler/test_poisson_posson.py Co-authored-by: Copilot <[email protected]> --------- Co-authored-by: Copilot <[email protected]>
* Introduction of the input structure for endpoints * Documentation + foundation of topology input * Improve Step validation, added unit test for the input * improved documentation for the input
* Improved input structure and pytest * Improved pytest structure accordingly to the new schema * definition of the metrics to be measured and update of the simulation input * improved documentations added rationale behind metrics * improved pytest logic and code coherence * Update src/app/core/helpers.py Co-authored-by: Copilot <[email protected]> --------- Co-authored-by: Copilot <[email protected]>
* definition of state and RqsGeneratorRuntime * defined edge runtime and more central logic for sampler * minor changes * minor changes * minor bug fixed * pytest adapted to the new structure, added pytest for rqs_state * Update src/app/core/runtime/rqs_generator.py Co-authored-by: Copilot <[email protected]> * Update src/app/core/runtime/rqs_generator.py Co-authored-by: Copilot <[email protected]> * Update src/app/core/runtime/edge.py Co-authored-by: Copilot <[email protected]> * Update src/app/core/runtime/edge.py Co-authored-by: Copilot <[email protected]> * Update tests/unit/runtime/test_requests_generator.py Co-authored-by: Copilot <[email protected]> * Update src/app/core/event_samplers/common_helpers.py Co-authored-by: Copilot <[email protected]> * Update src/app/core/event_samplers/common_helpers.py Co-authored-by: Copilot <[email protected]> * Update src/app/core/event_samplers/common_helpers.py Co-authored-by: Copilot <[email protected]> * minor changes --------- Co-authored-by: Copilot <[email protected]>
* defined client runtime + tests * initiated the server runtime plus correction of tghe structure * code refactor + test added for runtime + documentation * Update tests/conftest.py Co-authored-by: Copilot <[email protected]> --------- Co-authored-by: Copilot <[email protected]>
* defined architecture for the central collector + documentation * server modification to collect metrics and updated docs * Completed metric collector for server plus test * remove web app dependencies and added metrics for rqs latency * improved docs and improved metric collection * changes to make the code compatible with new changes * improved ci against toml changes * update lock file * minor change
* introduction to the analyzer class + tests * refactor of the analyzer
* Added LB structure in the input for the simulation + tests * introduced LB runtime + documentation * Added test for LB and small refactor to define the SimRunner
* added the first method to the simulation runner * added methods to handle nodes * completed simulation runtime + integration test
* new integration test + bug fixing * added integration and unit test * first tests working * minor change * first working example, upgrade docs, upgrade docstrings * improvements * path bug fixed
* Changed name from fastsim to asyncflow app folder renamed in asyncflow * improved docs
* Reafctor for the folder schemas + defined import for public api * Defined public api, improved docs
* Readme final, docs improved * Docs improvement, refactor analyzer, example added * Added scripts for setup, added pytest, improved readme * Example LB improvements * CI for main + system tests
# Conflicts: # pyproject.toml
# Conflicts: # pyproject.toml
* Structure defined for the pydantic model representing Event to be injected * Pydantic validation in the payload for events part 1 * Definition of the input for the events injection part 2 * minor fixes ruff and mypy compliant * full tests and docs added for input event injection * fixed minor bugs
* Events injection for edges part 1 * Logic + docs for the event injection to simulate a server down * Added pydantic validation + unit + int tests for eventinjection * added exaple for event inj yaml + builder added int tests * improved server model + system tests for eventinjection * ruff small fix
* new readme, ci to have pycov badge, comment on server runtime * small changes
# Conflicts: # .github/workflows/ci-main.yml # README.md # tests/system/test_sys_ev_inj_lb_two_servers.py
* added metric for server, added logic to collect them + tests * add kpi to collect in settings default * added metric plot for server and tests * Sweep introduction + MM1 heory vs observation + tests * analysis for sweep results + tests * mm1 notebook complete + server latency
… and end-to-end comparison reports (#26) * mmc split + notebook mmc split + random lb algo * tests for mmc
* changing api per ArrivalsGenerator and implementing distrib for ArrivalRuntime * Reafctoring API Arrival generator, tests integrations now working * Api Arrival generator migration complete with tests
Codecov Report❌ Patch coverage is 📢 Thoughts on this report? Let us know! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
What we changed
Rewrote LB FCFS to a token-based FIFO:
LoadBalancerRuntime._free_edges:simpy.Storeholding one token per free LB→server edge.get()s one token (blocking if none), then forwards the request to that edge.notify_server_free → lb.mark_free(edge_id)to return the token when they finish.Added
LoadBalancerRuntime.on_edge_added(edge_id)so EventInjection can re-enable edges cleanly (push a token when a server comes back)._lb_waiting_timeand exposes them vialb_waiting_timesLOAD_BALANCERentry.yield env.timeout(0)) to preserve causal order of timestamps.SimulationRunner.run()now passes the LB runtime intoResultsAnalyzer(lb=self._lb_runtime).ResultsAnalyzerexposesget_lb_waiting_times()and uses LB reference safely.MMc analyzer:
EventInjectionRuntimere-adds edges (server up) and callslb_on_edge_added(edge_id)(wired from the runner) so the token is immediately available again.OrderedDict; any stale tokens are discarded by the LB loop.New FCFS tests covering the token FIFO:
Adjusted annotations (mypy) and lints (no blind
except).