Skip to navigation

Chakra Replay

View as Markdown

Chakra Replay workload (test_template_name is ChakraReplay) replays execution traces from the Chakra execution trace format for performance analysis and debugging.

Usage Examples

Test TOML example:

name = "my_chakra_test"
description = "Example Chakra replay test"
test_template_name = "ChakraReplay"
[cmd_args]
trace_path = "/path/to/trace.et"

Test Scenario example:

name = "chakra-replay-test"
[[Tests]]
id = "chakra.1"
num_nodes = 1
time_limit = "00:10:00"
test_name = "my_chakra_test"

Test-in-Scenario example:

name = "chakra-replay-test"
[[Tests]]
id = "chakra.1"
num_nodes = 1
time_limit = "00:10:00"
name = "my_chakra_test"
description = "Example Chakra replay test"
test_template_name = "ChakraReplay"
[Tests.cmd_args]
trace_path = "/path/to/trace.et"

API Documentation

Command Arguments

class cloudai.workloads.chakra_replay.chakra_replay.ChakraReplayCmdArgs(

*,

docker_image_url: str,

mpi: str = ‘pmix’,

trace_type: str = ‘et’,

trace_path: str | None = None,

num_replays: int = 1,

**extra_data: Any,

)

Bases: CmdArgs

ChakraReplay test command arguments.

Test Definition

class cloudai.workloads.chakra_replay.chakra_replay.ChakraReplayTestDefinition(

*,

name: str,

description: str,

test_template_name: str,

cmd_args: ChakraReplayCmdArgs,

extra_env_vars: dict[str, str | List[str]] = {},

extra_cmd_args: dict[str, str] = {},

extra_container_mounts: list[str] = [],

git_repos: list[GitRepo] = [],

nsys: NsysConfiguration | None = None,

predictor: 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, Any] | None = None,

)

Bases: TestDefinition

Test object for ChakraReplay.