UCC#

This workload (test_template_name is UCCTest) allows users to execute UCC benchmarks within the CloudAI framework.

Usage Examples#

Test TOML example:

name = "ucc"
description = "Example UCC test"
test_template_name = "UCCTest"

[cmd_args]
docker_image_url = "nvcr.io#nvidia/pytorch:25.06-py3"

Test Scenario example:

name = "ucc-test"

[[Tests]]
id = "ucc.1"
num_nodes = 1
time_limit = "00:02:00"

test_name = "ucc"

Test-in-Scenario example:

name = "ucc-test"

[[Tests]]
id = "ucc.1"
num_nodes = 1
time_limit = "00:02:00"

name = "ucc"
description = "Example UCC test"
test_template_name = "UCCTest"

[Tests.cmd_args]
docker_image_url = "nvcr.io#nvidia/pytorch:25.06-py3"

API Documentation#

Command Arguments#

class cloudai.workloads.ucc_test.ucc.UCCCmdArgs(

*,

docker_image_url: str,

collective: Literal['allgather', 'allgatherv', 'allreduce', 'alltoall', 'alltoallv', 'barrier', 'bcast', 'gather', 'gatherv', 'reduce', 'reduce_scatter', 'reduce_scatterv', 'scatter', 'scatterv', 'memcpy', 'reducedt', 'reducedt_strided'] | list[Literal['allgather', 'allgatherv', 'allreduce', 'alltoall', 'alltoallv', 'barrier', 'bcast', 'gather', 'gatherv', 'reduce', 'reduce_scatter', 'reduce_scatterv', 'scatter', 'scatterv', 'memcpy', 'reducedt', 'reducedt_strided']] = 'alltoall',

b: int | list[int] = 1,

e: str | list[str] = '8M',

gen: str | list[str] | None = None,

use_deepep_matrix: bool = False,

**extra_data: Any,

)[source]#

Bases: CmdArgs

UCC test command arguments.

Test Definition#

class cloudai.workloads.ucc_test.ucc.UCCTestDefinition(*, name: str, description: str, test_template_name: str, cmd_args: ~cloudai.workloads.ucc_test.ucc.UCCCmdArgs, dse_excluded_args: list[str] = <factory>, extra_env_vars: dict[str, str | ~typing.List[str]] = {}, extra_cmd_args: dict[str, str] = {}, extra_container_mounts: list[str] = [], git_repos: list[~cloudai._core.installables.git_repo.GitRepo] = [], nsys: ~cloudai.models.workload.NsysConfiguration | None = None, predictor: ~cloudai.models.workload.PredictorConfig | None = None, agent: str = 'grid_search', agent_steps: int = 1, agent_metrics: list[str] = ['default'], agent_reward_function: str = 'inverse', agent_config: dict[str, ~typing.Any] | None = None, env_params: dict[str, ~cloudai.configurator.env_params.EnvParamSpec] = <factory>)[source]#

Bases: TestDefinition

Test object for UCC.

property is_domain_randomization_enabled: bool#

at least one env_params annotation.

Type:: Whether the config declares domain randomization

is_dse_excluded_arg(path: str) → bool#: Return whether a dot-separated cmd_args path should be ignored by DSE.

is_env_sampled(cmd_args_path: str) → bool#: Whether a cmd_args field is env-sampled (env draws it per trial, not the agent).

validator validate_env_params » all fields#

Validate env_params annotations against cmd_args.

env_params is an annotation: each key names a cmd_args field whose value is the candidate set (the single source of truth), and the entry carries only how to sample. So each key must name a real cmd_args field whose value is a candidate list; a scalar is already fixed, so annotating it is a meaningless label and is rejected here. When weights are declared, the list needs >= 2 values and the weights must align 1:1 with it. Sampling, persistence, the per-trial cmd_args overlay, and the cache key all live in CloudAIGymEnv; keeping this shape check in core lets the overlay stay agent- and workload-agnostic rather than re-implemented per workload.