oned::DirectCut< T, Pr > Class Template Reference

Greedily allocates to each processor the smallest interval with load less than average load. More...

#include <oned/direct_cut.hpp>

---

Detailed Description

template<typename T, typename Pr>
class oned::DirectCut< T, Pr >

Greedily allocates to each processor the smallest interval with load less than average load.

This algorithm was introduced in "Serge Miguet and Jean-Marc Pierson, Heuristics for 1D Rectilinear Partitioning as a Low Cost and High Quality Answer to Dynamic Load Balancing, HPCN Europe 1997" under the name H1. It sets cut points where B=P_iter*(Wtot/P).

Parameters:

	T	data type of weight of instance
	Pr	data type of prefixSumArray

List of all members.

Static Public Member Functions
static T	direct_cut (int procCount, const Pr &prefixSumArray, int length, int *cutIndexes, T)
	Applies direct cut algorithm to a given prefixSumArray.
Static Private Member Functions
static T	direct_cut_internal (int procCount, const Pr &prefixSumArray, int length, int *cutIndexes)

---

Member Function Documentation

template<typename T , typename Pr >

T oned::DirectCut< T, Pr >::direct_cut	(	int	procCount,
		const Pr &	prefixSumArray,
		int	length,
		int *	cutIndexes,
		T
	)			[static]

Applies direct cut algorithm to a given prefixSumArray.

Parameters:

[in]	procCount	is the total number of processors
[in]	prefixSumArray	always begins with 0 as first element.
[in]	length	is the exact size of prefixSumArray
[out]	cutIndexes	(must be allocated before calling!)

Returns:

Load of the max. loaded processor

template<typename T , typename Pr >

T oned::DirectCut< T, Pr >::direct_cut_internal	(	int	procCount,
		const Pr &	prefixSumArray,
		int	length,
		int *	cutIndexes
	)			[static, private]

---

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

• include/oned/direct_cut.hpp
• include/oned/direct_cut_impl.hpp