Files
cleveragents-core/benchmarks/subplan_execution_bench.py
T
freemo 6519f140a9
CI / benchmark-publish (pull_request) Has been skipped
CI / lint (pull_request) Successful in 15s
CI / build (pull_request) Successful in 18s
CI / quality (pull_request) Successful in 18s
CI / typecheck (pull_request) Successful in 34s
CI / security (pull_request) Successful in 48s
CI / unit_tests (pull_request) Successful in 2m16s
CI / docker (pull_request) Successful in 39s
CI / integration_tests (pull_request) Successful in 3m0s
CI / coverage (pull_request) Successful in 3m58s
CI / lint (push) Successful in 13s
CI / build (push) Successful in 15s
CI / quality (push) Successful in 17s
CI / typecheck (push) Successful in 31s
CI / benchmark-regression (push) Has been skipped
CI / security (push) Successful in 35s
CI / unit_tests (push) Successful in 3m23s
CI / docker (push) Successful in 38s
CI / integration_tests (push) Successful in 4m7s
CI / coverage (push) Successful in 4m48s
CI / benchmark-publish (push) Successful in 14m14s
CI / benchmark-regression (pull_request) Successful in 24m27s
feat(context): add hot/warm/cold tiers and actor views
Implement hot/warm/cold context tiers with ContextTier, ActorRole,
TieredFragment, TierBudget, ActorContextView, TierMetrics, and
ScopedBackendView models. ContextTierService provides store/get,
promotion/demotion with cold-tier summarisation hook, LRU eviction,
per-actor filtered views (strategist/executor/reviewer), and
project-scoped isolation via ScopedBackendView.

Settings: context_max_tokens_hot, context_max_decisions_warm,
context_max_decisions_cold. DI-wired as singleton context_tier_service.

Includes Behave BDD scenarios (30), Robot Framework integration tests (7),
ASV benchmarks (5 suites), and docs/reference/context_tiers.md.

Closes #208
2026-03-03 17:01:57 +00:00

156 lines
5.7 KiB
Python

"""Airspeed Velocity benchmarks for subplan execution scheduler overhead.
Measures scheduling, merge, and integrated execution+merge performance for
SubplanExecutionService and SubplanMergeService across all execution modes
and merge strategies.
"""
from __future__ import annotations
from cleveragents.application.services.subplan_execution_service import (
SubplanExecutionOutput,
SubplanExecutionService,
)
from cleveragents.application.services.subplan_merge_service import (
SubplanMergeService,
)
from cleveragents.domain.models.core.plan import (
ExecutionMode,
SubplanConfig,
SubplanMergeStrategy,
SubplanStatus,
)
_S1 = "01HGZ6FE0AQDYTR4BXVQZ6EA00"
_S2 = "01HGZ6FE0AQDYTR4BXVQZ6EB00"
_S3 = "01HGZ6FE0AQDYTR4BXVQZ6EC00"
def _make_status(subplan_id: str) -> SubplanStatus:
return SubplanStatus(
subplan_id=subplan_id,
action_name="local/bench-sub",
)
def _noop_executor(status: SubplanStatus) -> SubplanExecutionOutput:
return SubplanExecutionOutput(
subplan_id=status.subplan_id,
success=True,
files={f"src/{status.subplan_id[-4:]}.py": f"# {status.subplan_id}\n"},
files_changed=1,
)
class SubplanExecutionSchedulerSuite:
"""Benchmark SubplanExecutionService scheduling overhead."""
def setup(self) -> None:
"""Prepare fixtures for scheduler benchmarks."""
self.statuses = [_make_status(_S1), _make_status(_S2), _make_status(_S3)]
self.seq_config = SubplanConfig(execution_mode=ExecutionMode.SEQUENTIAL)
self.par_config = SubplanConfig(
execution_mode=ExecutionMode.PARALLEL, max_parallel=3
)
self.dep_config = SubplanConfig(execution_mode=ExecutionMode.DEPENDENCY_ORDERED)
self.dep_graph = {_S1: [], _S2: [_S1], _S3: [_S2]}
def time_sequential_execution(self) -> None:
"""Time sequential execution of 3 subplans."""
service = SubplanExecutionService(
config=self.seq_config, executor_fn=_noop_executor
)
service.execute_all(subplan_statuses=self.statuses, base_files={})
def time_parallel_execution(self) -> None:
"""Time parallel execution of 3 subplans."""
service = SubplanExecutionService(
config=self.par_config, executor_fn=_noop_executor
)
service.execute_all(subplan_statuses=self.statuses, base_files={})
def time_dependency_ordered_execution(self) -> None:
"""Time dependency-ordered execution of 3 subplans."""
service = SubplanExecutionService(
config=self.dep_config, executor_fn=_noop_executor
)
service.execute_all(
subplan_statuses=self.statuses,
base_files={},
dependency_graph=self.dep_graph,
)
def time_service_construction(self) -> None:
"""Time constructing SubplanExecutionService."""
SubplanExecutionService(config=self.seq_config, executor_fn=_noop_executor)
class SubplanMergeStrategySuite:
"""Benchmark SubplanMergeService strategy overhead."""
def setup(self) -> None:
"""Prepare fixtures for merge benchmarks."""
self.base_files = {"src/main.py": "line1\nline2\nline3\n"}
self.outputs = [
(_S1, {"src/main.py": "line1\nline2\nline3\nnew_a\n"}),
(_S2, {"src/main.py": "new_b\nline1\nline2\nline3\n"}),
]
self.overlapping_outputs = [
(_S1, {"src/main.py": "first version\n"}),
(_S2, {"src/main.py": "second version\n"}),
]
def time_git_three_way_merge(self) -> None:
"""Time git three-way merge of non-overlapping changes."""
service = SubplanMergeService(SubplanMergeStrategy.GIT_THREE_WAY)
service.merge(self.base_files, self.outputs)
def time_sequential_apply_merge(self) -> None:
"""Time sequential apply merge."""
service = SubplanMergeService(SubplanMergeStrategy.SEQUENTIAL_APPLY)
service.merge(self.base_files, self.overlapping_outputs)
def time_last_wins_merge(self) -> None:
"""Time last-wins merge."""
service = SubplanMergeService(SubplanMergeStrategy.LAST_WINS)
service.merge(self.base_files, self.overlapping_outputs)
def time_merge_service_construction(self) -> None:
"""Time constructing SubplanMergeService."""
SubplanMergeService(SubplanMergeStrategy.GIT_THREE_WAY)
def time_single_file_merge(self) -> None:
"""Time merging a single file output."""
service = SubplanMergeService(SubplanMergeStrategy.LAST_WINS)
service.merge(
{"src/main.py": "base\n"},
[(_S1, {"src/main.py": "modified\n"})],
)
class SubplanIntegrationSuite:
"""Benchmark integrated execution+merge overhead."""
def setup(self) -> None:
"""Prepare fixtures for integration benchmarks."""
self.statuses = [_make_status(_S1), _make_status(_S2)]
def time_sequential_with_last_wins(self) -> None:
"""Time sequential execution with last-wins merge."""
config = SubplanConfig(
execution_mode=ExecutionMode.SEQUENTIAL,
merge_strategy=SubplanMergeStrategy.LAST_WINS,
)
service = SubplanExecutionService(config=config, executor_fn=_noop_executor)
service.execute_all(subplan_statuses=self.statuses, base_files={})
def time_parallel_with_git_merge(self) -> None:
"""Time parallel execution with git three-way merge."""
config = SubplanConfig(
execution_mode=ExecutionMode.PARALLEL,
max_parallel=2,
merge_strategy=SubplanMergeStrategy.GIT_THREE_WAY,
)
service = SubplanExecutionService(config=config, executor_fn=_noop_executor)
service.execute_all(subplan_statuses=self.statuses, base_files={})