hebench::Utilities Namespace Reference

Classes
struct	BenchmarkRequest
struct	BenchmarkSession
class	BenchmarkConfigLoader
class	BenchmarkConfigBroker
	Provides facilities to configure the selection of benchmarks to run based on the supported workloads from a loaded backend. More...
class	RandomGenerator
class	TimingReportEx
class	ConfigExporterImpl
class	ConfigImporterImpl

Functions
std::string	convertToDirectoryName (const std::string &s, bool to_lowercase=true)
	Converts a string to directory name friendly. More...
void	printArraysAsColumns (std::ostream &os, const hebench::APIBridge::NativeDataBuffer **p_buffers, std::size_t count, hebench::APIBridge::DataType data_type, bool output_row_index=false, const char *separator=" ")
	Writes the collection of NativeDataBuffer as columns to the specified output stream. More...
template<typename T >
void	printArraysAsColumns (std::ostream &os, const hebench::APIBridge::NativeDataBuffer **p_buffers, std::size_t count, bool output_row_index=false, const char *separator=" ")
	Writes the collection of NativeDataBuffer as columns to the specified output stream. More...
template<>
std::string	ConfigImporterImpl::retrieveValue< std::string > (const std::string &s_value)

---

Class Documentation

hebench::Utilities::BenchmarkRequest

struct hebench::Utilities::BenchmarkRequest

Definition at line 29 of file hebench_config.h.

Collaboration diagram for hebench::Utilities::BenchmarkRequest:

Class Members
Configuration	configuration
size_t	index

hebench::Utilities::BenchmarkSession

struct hebench::Utilities::BenchmarkSession

Definition at line 35 of file hebench_config.h.

Collaboration diagram for hebench::Utilities::BenchmarkSession:

Class Members
vector< BenchmarkRequest >	benchmark_requests
vector< int8_t >	data
uint64_t	random_seed

Function Documentation

ConfigImporterImpl::retrieveValue< std::string >()

template<>

std::string hebench::Utilities::ConfigImporterImpl::retrieveValue< std::string > ( const std::string &  s_value )

inline

Definition at line 125 of file hebench_config.cpp.

convertToDirectoryName()

std::string hebench::Utilities::convertToDirectoryName	(	const std::string &	s,
		bool	to_lowercase = true
	)

Converts a string to directory name friendly.

Parameters

[in]	s	String to convert.
[in]	to_lowercase	Specifies whether to return all lowercase.

Returns

A string that can be used as a directory name.

This function will return a string generated from s where the result is composed only of alpha-numeric characters and non-repeating underscores. The result will have no leading or trailing underscores. If to_lowercase flag is true, all characters will lowercase, otherwise, they will keep the case originally encountered in s. Non-alpha-numeric characters will be turned into underscores.

For example:

s = "Hello, World!"
 
std::cout << convertToDirectoryName(s, true);
 
// Prints:
// hello_world

Definition at line 18 of file hebench_utilities_harness.cpp.

printArraysAsColumns() [1/2]

template<typename T >

void hebench::Utilities::printArraysAsColumns	(	std::ostream &	os,
		const hebench::APIBridge::NativeDataBuffer **	p_buffers,
		std::size_t	count,
		bool	output_row_index = false,
		const char *	separator = " "
	)

Writes the collection of NativeDataBuffer as columns to the specified output stream.

Parameters

	os	Stream where to write.
[in]	p_buffers	Array of NativeDataBuffer where each element will be output per column. Each element does not need to have the same number of rows. It is assumed that buffers in this array point to data of type T.
[in]	count	Number of elements in array p_buffers.
[in]	separator	Separator text to use when separating the columns.

The number of elements in each buffer pointed by p_buffers does not need to match. This method will print as many elements as each buffer has per row, so buffers with fewer elements will be skipped in the output after all their elements have been printed out.

Definition at line 17 of file hebench_utilities_harness.inl.

printArraysAsColumns() [2/2]

void hebench::Utilities::printArraysAsColumns	(	std::ostream &	os,
		const hebench::APIBridge::NativeDataBuffer **	p_buffers,
		std::size_t	count,
		hebench::APIBridge::DataType	data_type,
		bool	output_row_index = false,
		const char *	separator = " "
	)

Writes the collection of NativeDataBuffer as columns to the specified output stream.

Parameters

	os	Stream where to write.
[in]	p_buffers	Array of NativeDataBuffer where each element will be output per column. Each element does not need to have the same number of rows.
[in]	count	Number of elements in array p_buffers.
[in]	data_type	Type of data pointed to by p_buffers ' NativeDataBuffer.
[in]	separator	Separator text to use when separating the columns.

The number of elements in each buffer pointed by p_buffers does not need to match. This method will print as many elements as each buffer has per row, so buffers with fewer elements will be skipped in the output after all their elements have been printed out.

Definition at line 52 of file hebench_utilities_harness.cpp.