org.apache.sysds.runtime.compress.cost

Class MemoryCostEstimator

java.lang.Object

org.apache.sysds.runtime.compress.cost.MemoryCostEstimator

All Implemented Interfaces:

Serializable, ICostEstimate

public class MemoryCostEstimator
extends Object
implements ICostEstimate

See Also:

Serialized Form

Field Summary

Fields inherited from interface org.apache.sysds.runtime.compress.cost.ICostEstimate

LOG

Constructor Summary

Constructors

Constructor and Description
MemoryCostEstimator()

Method Summary

Modifier and Type	Method and Description
double	getCostOfCollectionOfGroups(Collection<CompressedSizeInfoColGroup> gss) Get the computation cost of a collection of column groups information.
double	getCostOfCollectionOfGroups(Collection<CompressedSizeInfoColGroup> gss, CompressedSizeInfoColGroup g)
double	getCostOfColumnGroup(CompressedSizeInfoColGroup g) If the instruction does not care about the inter column group cost, such as in memory cost or in computation cost of right or left matrix multiplication we simply estimate the cost of individual column groups.
double	getCostOfTwoGroups(CompressedSizeInfoColGroup g1, CompressedSizeInfoColGroup g2)
double	getUncompressedCost(int nRows, int nCols, int sparsity) Calculate the cost of execution if uncompressed.
boolean	isCompareAll() Function that returns true if the cost is based an an all compare all column groups false otherwise.
boolean	shouldAnalyze(CompressedSizeInfoColGroup g1, CompressedSizeInfoColGroup g2) Decide if the column groups should be analysed, or the worst case join should be expected.
boolean	shouldTryJoin(CompressedSizeInfoColGroup g1, CompressedSizeInfoColGroup g2) Decide if the column groups should try to join, this is to filter obviously bad joins out.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

MemoryCostEstimator

public MemoryCostEstimator()

Method Detail

getUncompressedCost

public double getUncompressedCost(int nRows,
                                  int nCols,
                                  int sparsity)

Description copied from interface: ICostEstimate

Calculate the cost of execution if uncompressed. This is used as an optimization goal to at least achieve, since if the uncompressed cost of calculation is lower than in compressed space we don't want to compress.

Specified by:

getUncompressedCost in interface ICostEstimate

Parameters:

nRows - Number of rows in compression

nCols - Number of columns in compression

sparsity - Sparsity of the input

Returns:

The uncompressed cost.

getCostOfColumnGroup

public double getCostOfColumnGroup(CompressedSizeInfoColGroup g)

Description copied from interface: ICostEstimate

If the instruction does not care about the inter column group cost, such as in memory cost or in computation cost of right or left matrix multiplication we simply estimate the cost of individual column groups.

Specified by:

getCostOfColumnGroup in interface ICostEstimate

Parameters:

g - Column group to estimate the cost of

Returns:

The Cost of this column group

getCostOfCollectionOfGroups

public double getCostOfCollectionOfGroups(Collection<CompressedSizeInfoColGroup> gss)

Description copied from interface: ICostEstimate

Get the computation cost of a collection of column groups information. Since a some operations execution time depends on all compare all calculating the cost needs all the current groups to consider. This method therefore provide a method of comparing cost of all compare all.

Specified by:

getCostOfCollectionOfGroups in interface ICostEstimate

Parameters:

gss - All the groups considered for this cost calculation

Returns:

The cost of the calculation

getCostOfCollectionOfGroups

public double getCostOfCollectionOfGroups(Collection<CompressedSizeInfoColGroup> gss,
                                          CompressedSizeInfoColGroup g)

Specified by:

getCostOfCollectionOfGroups in interface ICostEstimate

getCostOfTwoGroups

public double getCostOfTwoGroups(CompressedSizeInfoColGroup g1,
                                 CompressedSizeInfoColGroup g2)

Specified by:

getCostOfTwoGroups in interface ICostEstimate

isCompareAll

public boolean isCompareAll()

Description copied from interface: ICostEstimate

Function that returns true if the cost is based an an all compare all column groups false otherwise.

Specified by:

isCompareAll in interface ICostEstimate

Returns:

boolean

shouldAnalyze

public boolean shouldAnalyze(CompressedSizeInfoColGroup g1,
                             CompressedSizeInfoColGroup g2)

Description copied from interface: ICostEstimate

Decide if the column groups should be analysed, or the worst case join should be expected. This is use full if the column groups are very small and the in practice difference between joining or not is small.

Specified by:

shouldAnalyze in interface ICostEstimate

Parameters:

g1 - Group 1

g2 - Group 2

Returns:

If the joining should be analyzed.

shouldTryJoin

public boolean shouldTryJoin(CompressedSizeInfoColGroup g1,
                             CompressedSizeInfoColGroup g2)

Description copied from interface: ICostEstimate

Decide if the column groups should try to join, this is to filter obviously bad joins out.

Specified by:

shouldTryJoin in interface ICostEstimate

Parameters:

g1 - Group 1

g2 - Group 2

Returns:

Boolean specifying if they should try to join.