Expiration Timers

Expiration timers are set up on the same process as the agent and used from your script to deal with stuck workers. When you go into a code-block that has the potential to get stuck you can acquire an expiration timer, which instructs the timer server to kill the process if it does not release the timer by the self-imposed expiration deadline.


import torchelastic.timer as timer
import torchelastic.agent.server as agent

def main():
    start_method = "spawn"
    message_queue = mp.get_context(start_method).Queue()
    server = timer.LocalTimerServer(message, max_interval=0.01)
    server.start() # non-blocking

    spec = WorkerSpec(
    agent = agent.LocalElasticAgent(spec, start_method)

def trainer_func(message_queue):
    with timer.expires(after=60): # 60 second expiry
        # do some work

In the example above if trainer_func takes more than 60 seconds to complete, then the worker process is killed and the agent retries the worker group.

Client Methods


Configures a timer client. Must be called before using expires.

torch.distributed.elastic.timer.expires(after, scope=None, client=None)[source]

Acquires a countdown timer that expires in after seconds from now, unless the code-block that it wraps is finished within the timeframe. When the timer expires, this worker is eligible to be reaped. The exact meaning of “reaped” depends on the client implementation. In most cases, reaping means to terminate the worker process. Note that the worker is NOT guaranteed to be reaped at exactly + after, but rather the worker is “eligible” for being reaped and the TimerServer that the client talks to will ultimately make the decision when and how to reap the workers with expired timers.


with expires(after=10):

Server/Client Implementations

Below are the timer server and client pairs that are provided by torchelastic.


Timer server and clients always have to be implemented and used in pairs since there is a messaging protocol between the server and client.

class torch.distributed.elastic.timer.LocalTimerServer(mp_queue, max_interval=60, daemon=True)[source]

Server that works with LocalTimerClient. Clients are expected to be subprocesses to the parent process that is running this server. Each host in the job is expected to start its own timer server locally and each server instance manages timers for local workers (running on processes on the same host).

class torch.distributed.elastic.timer.LocalTimerClient(mp_queue)[source]

Client side of LocalTimerServer. This client is meant to be used on the same host that the LocalTimerServer is running on and uses pid to uniquely identify a worker. This is particularly useful in situations where one spawns a subprocess (trainer) per GPU on a host with multiple GPU devices.

Writing a custom timer server/client

To write your own timer server and client extend the torch.distributed.elastic.timer.TimerServer for the server and torch.distributed.elastic.timer.TimerClient for the client. The TimerRequest object is used to pass messages between the server and client.

class torch.distributed.elastic.timer.TimerRequest(worker_id, scope_id, expiration_time)[source]

Data object representing a countdown timer acquisition and release that is used between the TimerClient and TimerServer. A negative expiration_time should be interpreted as a “release” request.


the type of worker_id is implementation specific. It is whatever the TimerServer and TimerClient implementations have on to uniquely identify a worker.

class torch.distributed.elastic.timer.TimerServer(request_queue, max_interval, daemon=True)[source]

Entity that monitors active timers and expires them in a timely fashion. This server is responsible for reaping workers that have expired timers.

abstract clear_timers(worker_ids)[source]

Clears all timers for the given worker_ids.

abstract get_expired_timers(deadline)[source]

Returns all expired timers for each worker_id. An expired timer is a timer for which the expiration_time is less than or equal to the provided deadline.

abstract register_timers(timer_requests)[source]

Processes the incoming timer requests and registers them with the server. The timer request can either be a acquire-timer or release-timer request. Timer requests with a negative expiration_time should be interpreted as a release-timer request.

class torch.distributed.elastic.timer.TimerClient[source]

Client library to acquire and release countdown timers by communicating with the TimerServer.

abstract acquire(scope_id, expiration_time)[source]

Acquires a timer for the worker that holds this client object given the scope_id and expiration_time. Typically registers the timer with the TimerServer.

abstract release(scope_id)[source]

Releases the timer for the scope_id on the worker this client represents. After this method is called, the countdown timer on the scope is no longer in effect.


Access comprehensive developer documentation for PyTorch

View Docs


Get in-depth tutorials for beginners and advanced developers

View Tutorials


Find development resources and get your questions answered

View Resources