"""Scheduler for rollout tasks."""
import asyncio
import re
import time
import traceback
from collections import defaultdict, deque
from dataclasses import dataclass, field, replace
from typing import Dict, List, Optional, Tuple, Union
import ray
from trinity.common.config import Config
from trinity.common.experience import Experience
from trinity.common.models import InferenceModel
from trinity.common.workflows import Task
from trinity.explorer.workflow_runner import Status, WorkflowRunner
from trinity.utils.log import get_logger
[docs]
@dataclass
class TaskWrapper:
"""A wrapper for a task.
Each task can run multiple times (repeat_times) on same or different runners.
"""
task: Task
batch_id: Union[int, str]
sub_task_num: int = 1 # number of sub tasks splitted from this task
# if max_repeat_times_per_runner is set, one task may be splitted into multiple sub tasks
results: List[Tuple[Status, List[Experience]]] = field(default_factory=list)
[docs]
def calculate_task_level_metrics(metrics: List[Dict], is_eval: bool) -> Dict[str, float]:
"""Calculate task level metrics (mean) from multiple runs of the same task.
Args:
metrics (`List[Dict]`): A list of metric dictionaries from multiple runs of the same task.
is_eval (`bool`): Whether this is an evaluation task.
Returns:
`Dict[str, float]`: A dictionary of aggregated metrics, where each metric is averaged over all runs.
"""
if not metrics:
return {}
aggregated_metrics: Dict[str, List[float]] = defaultdict(list)
for m in metrics:
for key, value in m.items():
if isinstance(value, (int, float)):
aggregated_metrics[key].append(value)
if is_eval:
result = {}
for key, values in aggregated_metrics.items():
if "time/task_execution" in key or "time/run_execution" in key:
result[key] = sum(values) / len(values)
continue
k_list = []
k = 2
while k < len(values):
k_list.append(k)
k *= 2
k_list.append(len(values))
for k in k_list:
result[f"{key}/mean@{k}"] = sum(values[:k]) / k
result[f"{key}/best@{k}"] = max(values[:k])
result[f"{key}/worst@{k}"] = min(values[:k])
return result
else:
return {
key: sum(values) / len(values) for key, values in aggregated_metrics.items() if values
}
[docs]
class RunnerWrapper:
"""A wrapper for a WorkflowRunner"""
[docs]
def __init__(
self,
runner_id: int,
rollout_model: InferenceModel,
auxiliary_models: List[InferenceModel],
config: Config,
):
self.logger = get_logger(__name__)
self.runner_id = runner_id
self.rollout_model = rollout_model
self.auxiliary_models = auxiliary_models
self.config = config
self.retry_times = config.explorer.max_retry_times
self.timeout = config.explorer.max_timeout
self.namespace = ray.get_runtime_context().namespace
self.runner = self._create_runner()
self.state = {}
def _create_runner(self):
return (
ray.remote(WorkflowRunner)
.options(
num_cpus=0,
namespace=self.namespace,
scheduling_strategy="SPREAD",
runtime_env={
"env_vars": self.config.explorer.env_vars,
},
)
.remote(self.config, self.rollout_model, self.auxiliary_models, self.runner_id)
)
[docs]
async def prepare(self):
await self.runner.prepare.remote()
[docs]
async def update_state(self) -> None:
"""Get the runner state."""
self.state = await self.runner.get_runner_state.remote()
[docs]
async def run_with_retry(
self, task: TaskWrapper, repeat_times: int, run_id_base: int, timeout: float
) -> Tuple[Status, List, int, float]:
"""
Args:
task (`TaskWrapper`): The task to run.
repeat_times (`int`): The number of times to repeat the task.
run_id_base (`int`): The base run id for this task runs.
timeout (`float`): The timeout for each task run.
Returns:
`Status`: The return status of the task.
`List`: The experiences generated by the task.
`int`: The runner_id of current runner.
`float`: The time taken to run the task.
"""
last_exception_msg = None
await self.runner.__ray_ready__.remote()
start_time = time.time()
status = Status(ok=False, metrics=list())
exps = []
task2run = replace(
task.task,
rollout_args=replace(
task.task.rollout_args,
n=repeat_times,
),
)
try:
for attempt in range(self.retry_times + 1):
try:
status, exps = await asyncio.wait_for(
self.runner.run_task.remote(
task=task2run,
repeat_times=repeat_times,
run_id_base=run_id_base,
),
timeout=timeout,
)
if status.ok:
break
else:
self.logger.error(status.message)
except asyncio.TimeoutError:
last_exception_msg = f"Timeout when running task of batch {task.batch_id} at runner {self.runner_id} at attempt {attempt + 1}: {task.task}"
self.logger.error(last_exception_msg)
status = Status(ok=False, metrics=list(), message=last_exception_msg)
except Exception:
last_exception_msg = traceback.format_exc()
self.logger.warning(
f"Task execution attempt {attempt + 1} failed:\n{last_exception_msg}"
)
status = Status(ok=False, metrics=list(), message=last_exception_msg)
finally:
end_time = time.time()
status.metrics.append({"time/task_execution": end_time - start_time})
return status, exps, self.runner_id, end_time - start_time
[docs]
async def restart_runner(self):
old_runner = self.runner
self.runner = self._create_runner()
await self.runner.prepare.remote()
try:
ray.kill(old_runner)
except Exception:
pass
[docs]
def sort_batch_id(batch_id: Union[int, str]):
"""Priority of batch_id"""
# TODO: avoid sort the batch_id every time
if isinstance(batch_id, int):
return (batch_id, 0)
else:
match = re.match(r"^(\d+)", batch_id)
if match:
num = int(match.group(1))
return (num, 1)
else:
return (float("inf"), 1)
[docs]
class Scheduler:
"""Scheduler for rollout tasks.
Supports scheduling tasks to multiple runners, retrying failed tasks,
and collecting results at different levels.
"""
[docs]
def __init__(
self,
config: Config,
rollout_model: List[InferenceModel],
auxiliary_models: Optional[List[List[InferenceModel]]] = None,
):
self.logger = get_logger(__name__)
self.config = config
self.rollout_model = rollout_model
self.auxiliary_models = auxiliary_models or []
self.namespace = ray.get_runtime_context().namespace
self.default_timeout = config.explorer.max_timeout * (config.explorer.max_retry_times + 1)
self.max_retry_times = config.explorer.max_retry_times
self.max_repeat_times = config.explorer.max_repeat_times_per_runner
self.default_batch_size = config.buffer.batch_size
self.running = False
self.runner_num = len(rollout_model) * config.explorer.runner_per_model
self.runners: Dict[int, RunnerWrapper] = dict()
self.idle_runners = set() # runner_id of idle runners
self.busy_runners = dict() # runner_id -> task
self.pending_tasks: Dict[Union[int, str], deque] = defaultdict(
deque
) # batch_id -> (task, repeat_times, run_id_base)
self.running_tasks: Dict[Union[int, str], set[asyncio.Future]] = defaultdict(
set
) # batch_id -> futures
self.task_num_map: Dict[Union[int, str], int] = defaultdict(
int
) # batch_id -> tasks scheduled under this batch_id
self.running_task_map: Dict[asyncio.Future, TaskWrapper] = dict() # future -> task
self.completed_tasks: Dict[
Union[int, str], deque[Tuple[Status, List[Experience]]]
] = defaultdict(
deque
) # batch_id -> results
self.scheduler_task: Optional[asyncio.Task] = None
self.running = False
self.total_running_time = 0.0
self.total_completed_tasks = 0
async def _create_runner(
self,
runner_id: int,
):
runner = RunnerWrapper(
runner_id=runner_id,
rollout_model=self.rollout_model[runner_id % len(self.rollout_model)],
auxiliary_models=[
self.auxiliary_models[j][runner_id % len(self.auxiliary_models[j])]
for j in range(len(self.auxiliary_models))
],
config=self.config,
)
await runner.prepare()
self.runners[runner_id] = runner
self.idle_runners.add(runner_id)
async def _restart_runner(self, runner_id: int):
"""Restart a runner."""
await self.runners[runner_id].restart_runner()
if runner_id in self.busy_runners:
task = self.busy_runners.pop(runner_id)
self.logger.warning(
f"Runner {runner_id} failed to run task at batch_id {task.batch_id}: {task.task.raw_task}"
)
self.idle_runners.add(runner_id)
self.logger.info(f"Runner {runner_id} restarted.")
async def _scheduler_loop(self) -> None:
self.logger.info("Scheduler loop started.")
while self.running:
try:
await self._schedule_pending_tasks()
await asyncio.sleep(0.01)
except Exception:
self.logger.error(f"Error in scheduler loop:\n{traceback.format_exc()}")
await asyncio.sleep(0.1)
self.logger.info("Scheduler loop stopped.")
async def _monitor_runner_state_loop(self) -> None:
interval = self.config.explorer.runner_state_report_interval
if interval <= 0:
self.logger.info("Runner state monitoring loop disabled.")
return
self.logger.info("Runner state monitoring loop started.")
while self.running:
try:
await asyncio.gather(*[runner.update_state() for runner in self.runners.values()])
self.print_all_state()
except Exception:
self.logger.error(
f"Error in runner state monitoring loop:\n{traceback.format_exc()}"
)
await asyncio.sleep(interval)
self.logger.info("Runner state monitoring loop stopped.")
async def _schedule_pending_tasks(self) -> None:
if not self.idle_runners:
return
# TODO: Support more advanced scheduling strategies
for batch_id in sorted(self.pending_tasks.keys(), key=sort_batch_id):
task_queue = self.pending_tasks[batch_id]
while task_queue and self.idle_runners:
task, repeat_times, run_id_base = task_queue.pop()
runner_id = self.idle_runners.pop()
self.busy_runners[runner_id] = task
future = asyncio.create_task(
self.runners[runner_id].run_with_retry(
task,
repeat_times=repeat_times,
run_id_base=run_id_base,
timeout=self.dynamic_timeout(),
)
)
self.running_task_map[future] = task
future.add_done_callback(self.task_done_callback)
self.running_tasks[batch_id].add(future)
if not task_queue:
del self.pending_tasks[batch_id]
[docs]
def task_done_callback(self, async_task: asyncio.Task):
task = self.running_task_map.pop(async_task)
if async_task.cancelled():
return
elif async_task.exception():
self.logger.error(f"Task {task.task.task_id} failed: {async_task.exception()}")
return
else:
status, exps, runner_id, run_time = async_task.result()
if not task.task.is_eval: # only count running time for non-eval tasks
self.total_running_time += run_time
self.total_completed_tasks += 1
task.results.append((status, exps))
self.busy_runners.pop(runner_id)
self.idle_runners.add(runner_id)
# If all sub runs in a task are completed
if len(task.results) == task.sub_task_num:
task_experiences = []
task_metrics = []
all_success = True
for s, exp in task.results:
task_metrics.extend(s.metrics)
task_experiences.extend(exp)
if not s.ok:
all_success = False
# calculate task level metrics
task_status = Status(
ok=all_success,
metrics=[calculate_task_level_metrics(task_metrics, task.task.is_eval)],
)
self.completed_tasks[task.batch_id].appendleft((task_status, task_experiences))
self.logger.debug(f"Task completed (batch_id {task.batch_id}).")
if task.batch_id in self.running_tasks:
self.running_tasks[task.batch_id].remove(async_task)
if not self.running_tasks[task.batch_id]:
del self.running_tasks[task.batch_id]
def _clear_timeout_tasks(self, batch_id: Union[int, str]) -> None:
if batch_id in self.pending_tasks:
self.logger.info(f"Clear timeout pending tasks at batch_id {batch_id}.")
del self.pending_tasks[batch_id]
if batch_id in self.running_tasks:
self.logger.info(f"Clear timeout running tasks at batch_id {batch_id}.")
for future in self.running_tasks[batch_id]:
future.cancel()
del self.running_tasks[batch_id]
self.task_num_map.pop(batch_id, None)
[docs]
async def start(self) -> None:
if self.running:
return
self.running = True
await asyncio.gather(*[self._create_runner(i) for i in range(self.runner_num)])
self.scheduler_task = asyncio.create_task(self._scheduler_loop())
ready_refs = [runner.runner.__ray_ready__.remote() for runner in self.runners.values()]
await asyncio.gather(*ready_refs)
self.monitor_task = asyncio.create_task(self._monitor_runner_state_loop())
self.logger.info(f"Starting Scheduler with {self.runner_num} runners")
[docs]
async def stop(self) -> None:
if not self.running:
return
self.running = False
all_running_futures = []
for futures in self.running_tasks.values():
all_running_futures.extend(futures)
if all_running_futures:
self.logger.info(f"Waiting for {len(all_running_futures)} running tasks to complete...")
await asyncio.gather(*all_running_futures, return_exceptions=True)
if self.scheduler_task:
self.scheduler_task.cancel()
try:
await self.scheduler_task
except asyncio.CancelledError:
pass
if self.monitor_task:
self.monitor_task.cancel()
try:
await self.monitor_task
except asyncio.CancelledError:
pass
self.logger.info("Scheduler stopped")
[docs]
def schedule(self, tasks: List[Task], batch_id: Union[int, str]) -> None:
"""Schedule the provided tasks.
Args:
tasks (`List[Task]`): The tasks to schedule.
batch_id (`Union[int, str]`): The id of provided tasks. It should be an integer or a string
starting with an integer (e.g., 123, "123/my_task")
"""
if not tasks:
return
self.task_num_map[batch_id] += len(tasks)
self._split_and_submit_tasks(tasks, batch_id=batch_id)
def _split_and_submit_tasks(self, tasks: List[Task], batch_id: Union[int, str]) -> None:
for i, task in enumerate(tasks):
assert task.repeat_times is not None, "Task repeat_times should not be None"
task_wrapper = TaskWrapper(
task=replace(task, batch_id=batch_id, task_id=i),
batch_id=batch_id,
)
if self.max_repeat_times is None:
task_wrapper.sub_task_num = 1
self.pending_tasks[batch_id].appendleft((task_wrapper, task.repeat_times, 0))
continue
sub_tasks = []
for run_id_base in range(0, task.repeat_times, self.max_repeat_times):
repeat_times = min(self.max_repeat_times, task.repeat_times - run_id_base)
sub_tasks.append((task_wrapper, repeat_times, run_id_base))
task_wrapper.sub_task_num = len(sub_tasks)
self.pending_tasks[batch_id].extendleft(sub_tasks)
[docs]
def dynamic_timeout(self, timeout: Optional[float] = None) -> float:
"""Calculate dynamic timeout based on historical data."""
max_timeout = timeout or self.default_timeout
if not self.config.explorer.dynamic_timeout.enable:
return max_timeout
if self.total_completed_tasks < self.default_batch_size:
return max_timeout
avg_time_per_task = self.total_running_time / self.total_completed_tasks
return min(
max_timeout,
avg_time_per_task * self.config.explorer.dynamic_timeout.ratio,
)
[docs]
async def get_results(
self,
batch_id: Union[int, str],
min_num: Optional[int] = None,
timeout: Optional[float] = None,
clear_timeout_tasks: bool = True,
) -> Tuple[List[Status], List[Experience]]:
"""Get the result of tasks at the specific batch_id.
Args:
batch_id (`Union[int, str]`): Only wait for tasks at this batch.
min_num (`int`): The minimum number of tasks to wait for. If `None`, wait for all tasks at `batch_id`.
timeout (`float`): The timeout for waiting for tasks to finish. If `None`, wait for default timeout.
clear_timeout_tasks (`bool`): Whether to clear timeout tasks.
"""
timeout = timeout or self.default_timeout
start_time = time.time()
scheduled_num = self.task_num_map.get(batch_id, 0)
if min_num is None:
min_num = scheduled_num
elif min_num > scheduled_num:
self.logger.warning(
f"Requested min_num {min_num} is greater than scheduled tasks {scheduled_num} at batch_id {batch_id}. Adjusting min_num to {scheduled_num}."
)
min_num = scheduled_num
self.logger.debug(f"Waiting for {min_num} tasks to complete...")
min_threshold_reached_time = None
while time.time() - start_time <= timeout:
completed_count = len(self.completed_tasks.get(batch_id, []))
if completed_count >= min_num:
min_threshold_reached_time = min_threshold_reached_time or time.time()
if (completed_count >= scheduled_num) or (
time.time() - min_threshold_reached_time
>= self.config.explorer.over_rollout.wait_after_min
):
break
await asyncio.sleep(0.1)
if time.time() - start_time > timeout:
self.logger.error(
f"Timed out waiting for tasks at batch {batch_id} to complete after {timeout} seconds"
)
if clear_timeout_tasks:
self._clear_timeout_tasks(batch_id=batch_id)
runners_to_restart = []
for runner_id, task in list(self.busy_runners.items()):
if task.batch_id == batch_id:
runners_to_restart.append(runner_id)
asyncio.gather(
*[self._restart_runner(runner_id) for runner_id in runners_to_restart]
)
statuses = []
experiences = []
completed_queue = self.completed_tasks.get(batch_id, deque())
while completed_queue:
status, exps = completed_queue.pop()
statuses.append(status)
if isinstance(exps, list):
experiences.extend(exps)
else:
experiences.append(exps)
if batch_id in self.completed_tasks and not self.completed_tasks[batch_id]:
del self.completed_tasks[batch_id]
completed_count = len(statuses)
if completed_count < min_num:
self.logger.warning(
f"Timeout reached, only {completed_count}/{min_num} tasks completed"
)
return statuses, experiences
[docs]
def has_step(self, batch_id: Union[int, str]) -> bool:
return (
batch_id in self.completed_tasks
or batch_id in self.pending_tasks
or batch_id in self.running_tasks
)
[docs]
async def wait_all(
self, timeout: Optional[float] = None, clear_timeout_tasks: bool = True
) -> None:
"""Wait for all tasks to complete without poping results. If timeout reached, raise TimeoutError.
Args:
timeout (`float`): timeout in seconds. Raise `TimeoutError` when no new tasks is completed within timeout.
clear_timeout_tasks (`bool`): Whether to clear timeout tasks.
"""
timeout = timeout or self.default_timeout
start_time = time.time()
self.logger.debug("Waiting for all tasks to complete...")
last_completed_count = 0
while time.time() - start_time < timeout:
has_pending = bool(self.pending_tasks)
has_running = bool(self.running_tasks)
if not has_pending and not has_running:
self.logger.debug("All tasks completed successfully")
return
completed_count = sum(len(tasks) for tasks in self.completed_tasks.values())
if completed_count != last_completed_count:
# flush timeout when new tasks are completed
start_time = time.time()
last_completed_count = completed_count
await asyncio.sleep(0.1)
pending_count = sum(len(tasks) for tasks in self.pending_tasks.values())
running_count = sum(len(futures) for futures in self.running_tasks.values())
error_msg = f"Timeout after {timeout} seconds. Still have {pending_count} pending tasks and {running_count} running tasks."
self.logger.error(error_msg)
if clear_timeout_tasks:
for batch_id in self.pending_tasks.keys() | self.running_tasks.keys():
self._clear_timeout_tasks(batch_id)
asyncio.gather(
*[self._restart_runner(runner_id) for runner_id in self.busy_runners.keys()]
)
raise TimeoutError(error_msg)
[docs]
def get_key_state(self, key: str) -> Dict:
"""Get the scheduler state.
Args:
key (`str`): The key of the state to get.
Returns:
`Dict`: A dictionary of runner ids to their state for the given key.
"""
result = {}
for runner in self.runners.values():
runner_state = runner.state
if runner_state and key in runner_state:
result[runner.runner_id] = runner_state[key]
return result
[docs]
def get_runner_state(self, runner_id: int) -> Dict:
"""Get the scheduler state.
Args:
runner_id (`int`): The id of the runner.
Returns:
`Dict`: The state of the runner.
"""
runner = self.runners.get(runner_id, None)
if runner:
return runner.state
else:
return {}
[docs]
def get_all_state(self) -> Dict:
"""Get all runners' state.
Returns:
`Dict`: The state of all runners.
"""
result = {}
for runner in self.runners.values():
runner_state = runner.state
if runner_state:
result[runner.runner_id] = runner_state
return result
[docs]
def print_all_state(self) -> None:
"""Print all runners' state in a clear, aligned table format."""
all_keys = set()
for runner in self.runners.values():
runner_state = runner.state
if runner_state:
all_keys.update(runner_state.keys())
all_keys = sorted(all_keys)
# Prepare header
header = ["runner_id"] + all_keys # type: ignore [operator]
# Prepare rows
rows = []
for runner in self.runners.values():
runner_state = runner.state or {}
row = [str(runner.runner_id)]
for key in all_keys:
value = runner_state.get(key, "-")
row.append(str(value))
rows.append(row)
# Calculate column widths
col_widths = [max(len(str(x)) for x in col) for col in zip(header, *rows)]
# Print header
header_line = " | ".join(str(h).ljust(w) for h, w in zip(header, col_widths))
self.logger.info(header_line)
self.logger.info("-+-".join("-" * w for w in col_widths))
# Print each row
for row in rows:
line = " | ".join(str(cell).ljust(w) for cell, w in zip(row, col_widths))
self.logger.info(line)
