stapl::skeletons::skeletons_impl::alltoall_pd< T, tags::hybrid > Class Template Reference

The hybrid alltoall parametric dependency is used for alltoall skeleton in the cases of having large messages. The parametric dependency in this case is defined as a hybrid of the flat and butterfly-based version of alltoall. More...

Public Member Functions
template<typename Coord , typename Visitor , typename In >
void	case_of (Coord const &skeleton_size, Coord const &coord, Visitor &visitor, In &&in_flow) const
	If coord is <i, h, ...> it wraps the WF with the following inputs and sends it to the visitor along with the m_wf. More...
template<typename Coord >
std::size_t	butterfly_size (Coord const &skeleton_size, Coord const &coord) const
template<typename Coord >
std::size_t	butterflied_index (Coord const &skeleton_size, Coord const &coord) const
template<typename Size , typename Coord , typename FlowIndex >
std::size_t	consumer_count (Size const &skeleton_size, Coord const &producer_coord, FlowIndex const &) const
	determines how many of the instances of this parametric dependency will be consuming from a producer with a given coordinate. This is a reverse query as compared to case_of. More...
template<typename Coord >
int	get_result_id (Coord const &, Coord const &) const
template<typename Coord , typename Span >
void	configure (Coord &&, Span &&)

Public Types
using	consumer_count_filter_type = skeletons::filters::filter< 2 >
using	op_type = alltoall_merge< T, tags::hybrid >

Static Public Attributes
static constexpr std::size_t	in_port_size = 1
static constexpr std::size_t	op_arity = 2

Detailed Description

template<typename T>
class stapl::skeletons::skeletons_impl::alltoall_pd< T, tags::hybrid >

The hybrid alltoall parametric dependency is used for alltoall skeleton in the cases of having large messages. The parametric dependency in this case is defined as a hybrid of the flat and butterfly-based version of alltoall.

In this case, there are log2(n) stages. In each stage, each participant at the upper half of a butterfly communicates with every participant in the lower half of the same butterfly, and vice versa.

Note

The dependencies at each step are permuted to avoid network pressure on a single node at the same time.

Member Function Documentation

case_of()

template<typename T >

template<typename Coord , typename Visitor , typename In >

void stapl::skeletons::skeletons_impl::alltoall_pd< T, tags::hybrid >::case_of	(	Coord const &	skeleton_size,
		Coord const &	coord,
		Visitor &	visitor,
		In &&	in_flow
	)		const

If coord is <i, h, ...> it wraps the WF with the following inputs and sends it to the visitor along with the m_wf.

• in<0>[0, ..., n]

Parameters

skeleton_size	<n, m, p, ...> where each element is potentially multi-dimensional.
coord	<i, j, k, ...> where i < n, j < m, k < p
visitor	the information about WF and input is passed so that later this information can be converted to a node in the dependence graph
in_flow	a tuple of input flows to consume from

consumer_count()

template<typename T >

template<typename Size , typename Coord , typename FlowIndex >

std::size_t stapl::skeletons::skeletons_impl::alltoall_pd< T, tags::hybrid >::consumer_count	(	Size const &	skeleton_size,
		Coord const &	producer_coord,
		FlowIndex const &
	)		const

determines how many of the instances of this parametric dependency will be consuming from a producer with a given coordinate. This is a reverse query as compared to case_of.

Parameters

skeleton_size	the size of this skeleton
producer_coord	the coordinate of the producer element which is providing data to this parametric dependency

Template Parameters

FlowIndex

the flow index on which this request is sent

---

The documentation for this class was generated from the following file:

• alltoall_hybrid_pd.hpp