Teuchos::PerformanceMonitorUtils Class Reference

Provides common capabilities for collecting and reporting performance data across processors. More...

#include <Teuchos_PerformanceMonitorUtils.hpp>

List of all members.

Static Public Member Functions
void	synchNames (const MPIComm &comm, const Array< string > &localNames, Array< string > &allNames)
	Synchronizes lists of metric names (e.g., timer names) across processors. This is necessary because some functions may not have been invoked on some processors, so that their named timers/counters aren't created on that processor. This function does a set union of all names created on all processors. It is called by synchValues().
void	synchValues (const MPIComm &comm, const Array< string > &localNames, const Array< Array< double > > &localValues, Array< string > &allNames, Array< Array< double > > &allValues)
	Creates zero values for metrics absent on this processor but present on other processors. This function uses a call to synchNames() to inform this processor of the existence of other metrics on other processors.
void	reduce (const MPIComm &comm, const EMetricReduction &reductionType, const Array< double > &localVals, Array< double > &reducedVals)
	Compute reduced performance metrics across processors, for example, min, max, or total times or flop counts.

---

Detailed Description

Provides common capabilities for collecting and reporting performance data across processors.

Definition at line 56 of file Teuchos_PerformanceMonitorUtils.hpp.

---

Member Function Documentation

void PerformanceMonitorUtils::synchNames (  const MPIComm &  comm, const Array< string > &  localNames, Array< string > &  allNames )  [static]

Synchronizes lists of metric names (e.g., timer names) across processors. This is necessary because some functions may not have been invoked on some processors, so that their named timers/counters aren't created on that processor. This function does a set union of all names created on all processors. It is called by synchValues(). Parameters: comm  [in] the communicator over which name lists are being synchronized localNames  [in] the names appearing on the local processor allNames  [out] the set union of name lists from all processors Definition at line 36 of file Teuchos_PerformanceMonitorUtils.cpp.

void PerformanceMonitorUtils::synchValues (  const MPIComm &  comm, const Array< string > &  localNames, const Array< Array< double > > &  localValues, Array< string > &  allNames, Array< Array< double > > &  allValues )  [static]

Creates zero values for metrics absent on this processor but present on other processors. This function uses a call to synchNames() to inform this processor of the existence of other metrics on other processors. Parameters: comm  [in] the communicator over which name lists are being synchronized localNames  [in] the names appearing on the local processor localValues  [in] the values appearing on the local processor allNames  [out] the set union of name lists from all processors allValues  [out] the metric values from all processors Definition at line 77 of file Teuchos_PerformanceMonitorUtils.cpp.

void PerformanceMonitorUtils::reduce (  const MPIComm &  comm, const EMetricReduction &  reductionType, const Array< double > &  localVals, Array< double > &  reducedVals )  [static]

Compute reduced performance metrics across processors, for example, min, max, or total times or flop counts. Parameters: comm  [in] The MPIComm object representing the communicator on which the reduction is to be done. reductionType  [in] the reduction operation to be performed localVals  [in] The metrics on this processor (after synchronization through a call to synchValues()) reducedVals  [out] The reduced metrics Definition at line 125 of file Teuchos_PerformanceMonitorUtils.cpp.

---

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

• Teuchos_PerformanceMonitorUtils.hpp
• Teuchos_PerformanceMonitorUtils.cpp

---