Configuration

Configuration objects passed to communicator creation methods, plus the flag enums they consume.

NCCLConfig

Used by Communicator.init(), Communicator.initialize(), Communicator.split(), Communicator.shrink(), and Communicator.grow(). Fields left unset (None) remain at NCCL’s internal default; values are validated by the C library when the config is consumed.

class nccl.core.NCCLConfig(*, blocking: bool | None = None, cga_cluster_size: int | None = None, min_ctas: int | None = None, max_ctas: int | None = None, net_name: str | None = None, split_share: bool | None = None, traffic_class: int | None = None, comm_name: str | None = None, collnet_enable: bool | None = None, cta_policy: CTAPolicy | None = None, shrink_share: bool | None = None, nvls_ctas: int | None = None, n_channels_per_net_peer: int | None = None, nvlink_centric_sched: bool | None = None, graph_usage_mode: int | None = None, num_rma_ctx: int | None = None, max_p2p_peers: int | None = None, graph_stream_ordering: int | None = None)

Bases: LowppSpec

NCCL configuration for communicator initialization.

Provides configuration options for NCCL communicators, allowing fine-tuning of performance and behavior characteristics. Fields not set in the constructor remain at NCCL’s internal default; values are validated by the C library when the config is consumed.

See also

ncclConfig_t for the description of each field.

blocking: bool | None = None

cga_cluster_size: int | None = None

collnet_enable: bool | None = None

comm_name: str | None = None

cta_policy: CTAPolicy | None = None

graph_stream_ordering: int | None = None

graph_usage_mode: int | None = None

max_ctas: int | None = None

max_p2p_peers: int | None = None

min_ctas: int | None = None

n_channels_per_net_peer: int | None = None

net_name: str | None = None

num_rma_ctx: int | None = None

nvlink_centric_sched: bool | None = None

nvls_ctas: int | None = None

shrink_share: bool | None = None

split_share: bool | None = None

traffic_class: int | None = None

CTAPolicy

class nccl.core.CTAPolicy(*values)

Bases: IntFlag

NCCL performance policy for CTA scheduling, used by NCCLConfig.cta_policy.

DEFAULT = 0

EFFICIENCY = 1

ZERO = 2

NCCLDevCommRequirements

Used by Communicator.create_dev_comm(). Fields left unset (None) remain at NCCL’s internal default.

class nccl.core.NCCLDevCommRequirements(*, lsa_multimem: bool | None = None, barrier_count: int | None = None, lsa_barrier_count: int | None = None, rail_gin_barrier_count: int | None = None, lsa_ll_a2a_block_count: int | None = None, lsa_ll_a2a_slot_count: int | None = None, gin_force_enable: bool | None = None, gin_context_count: int | None = None, gin_signal_count: int | None = None, gin_counter_count: int | None = None, gin_connection_type: NcclGinConnectionType | None = None, gin_exclusive_contexts: bool | None = None, gin_queue_depth: int | None = None, gin_traffic_class: int | None = None, world_gin_barrier_count: int | None = None, gin_strong_signals_required: bool | None = None, gin_va_signals_required: bool | None = None, teams: tuple[TeamRequirement, ...] = (), resources: tuple[LsaBarrierRequirement | GinBarrierRequirement | LLA2ARequirement, ...] = ())

Bases: LowppSpec

NCCL device communicator requirements configuration.

This is a reusable high-level Python request consumed by Communicator.create_dev_comm(). Per-team requirements are declared through the teams tuple. Each call snapshots the request into independent low-level ncclDevCommRequirements_t and linked ncclTeamRequirements_t storage, including separate multimem output handles. NCCL copies the requirements and linked-list nodes before the call returns; the resulting DevCommResource retains the storage referenced by each outMultimemHandle. This object may therefore be changed between calls without affecting device communicators that were already created. Do not mutate it concurrently with Communicator.create_dev_comm().

See also

ncclDevCommRequirements for the description of each field.

barrier_count: int | None = None

gin_connection_type: NcclGinConnectionType | None = None

gin_context_count: int | None = None

gin_counter_count: int | None = None

gin_exclusive_contexts: bool | None = None

gin_force_enable: bool | None = None

gin_queue_depth: int | None = None

gin_signal_count: int | None = None

gin_strong_signals_required: bool | None = None

gin_traffic_class: int | None = None

gin_va_signals_required: bool | None = None

lsa_barrier_count: int | None = None

lsa_ll_a2a_block_count: int | None = None

lsa_ll_a2a_slot_count: int | None = None

lsa_multimem: bool | None = None

rail_gin_barrier_count: int | None = None

resources: tuple[LsaBarrierRequirement | GinBarrierRequirement | LLA2ARequirement, ...] = ()

teams: tuple[TeamRequirement, ...] = ()

world_gin_barrier_count: int | None = None

Requirement entries

The element types of NCCLDevCommRequirements.teams and NCCLDevCommRequirements.resources.

TeamRequirement

class nccl.core.TeamRequirement(team: NCCLTeam, multimem: bool = False)

Bases: object

A per-team requirement for device communicator creation.

Pass a tuple of these as NCCLDevCommRequirements.teams. When multimem is True, NCCL allocates a multicast handle for the team, retrievable afterwards via multimem_handle().

multimem: bool = False

team: NCCLTeam

LsaBarrierRequirement

class nccl.core.LsaBarrierRequirement(team: NCCLTeam, n_barriers: int)

Bases: object

Requests an LSA barrier resource on team with n_barriers barriers.

Add to NCCLDevCommRequirements.resources; the finalized LsaBarrierHandle is returned in resource_handles.

team: NCCLTeam

n_barriers: int

GinBarrierRequirement

class nccl.core.GinBarrierRequirement(team: NCCLTeam, n_barriers: int)

Bases: object

Requests a GIN barrier resource on team with n_barriers barriers.

Add to NCCLDevCommRequirements.resources; the finalized GinBarrierHandle is returned in resource_handles.

team: NCCLTeam

n_barriers: int

LLA2ARequirement

class nccl.core.LLA2ARequirement(n_blocks: int, max_elements: int, max_element_size: int)

Bases: object

Requests a low-latency all-to-all resource with n_blocks blocks, sized to hold up to max_elements elements of at most max_element_size bytes each.

Add to NCCLDevCommRequirements.resources; the finalized LLA2AHandle is returned in resource_handles.

n_blocks: int

max_elements: int

max_element_size: int