Struct DualModuleParallelConfig 

pub struct DualModuleParallelConfig {
    pub thread_pool_size: usize,
    pub edges_in_fusion_unit: bool,
    pub enable_parallel_execution: bool,
}

Fields

thread_pool_size: usize

enable async execution of dual operations; only used when calling top-level operations, not used in individual units

edges_in_fusion_unit: bool

strategy of edges placement: if edges are placed in the fusion unit, it’s good for software implementation because there are no duplicate edges and no unnecessary vertices in the descendant units. On the other hand, it’s not very favorable if implemented on hardware: the fusion unit usually contains a very small amount of vertices and edges for the interfacing between two blocks, but maintaining this small graph may consume additional hardware resources and increase the decoding latency. I want the algorithm to finally work on the hardware efficiently so I need to verify that it does work by holding all the fusion unit’s owned vertices and edges in the descendants, although usually duplicated.

enable_parallel_execution: bool

enable parallel execution of a fused dual module

Trait Implementations

impl Clone for DualModuleParallelConfig

fn clone(&self) -> DualModuleParallelConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for DualModuleParallelConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for DualModuleParallelConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for DualModuleParallelConfig

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for DualModuleParallelConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.