Utilities for benchmarking hybrid vs fully distributed solvers. More...

#include <fstream>
#include <iomanip>
#include <iostream>
#include <mpi.h>
#include <sstream>
#include <string>
#include <vector>

Include dependency graph for utils.h:

This graph shows which files directly or indirectly include this file:

Classes
struct	BenchmarkUtils::BenchmarkResult
	Structure to hold benchmark results for a single run. More...
class	BenchmarkUtils::ResultCollector
	Class to collect and output benchmark results. More...
class	BenchmarkUtils::Timer
	High-resolution timer using MPI_Wtime for consistency. More...
struct	BenchmarkUtils::ScalingConfig
	Configuration for scaling tests. More...

Namespaces
namespace	BenchmarkUtils

Functions
double	BenchmarkUtils::compute_efficiency (const double t1, const double tp, int p)
	Compute parallel efficiency.
double	BenchmarkUtils::compute_speedup (const double t1, const double tp)
	Compute speedup.
double	BenchmarkUtils::get_process_memory_mb ()
	Get memory info from /proc/self/status (Linux specific)

Detailed Description

Utilities for benchmarking hybrid vs fully distributed solvers.

This header provides:

Classes