Apache
/
hadoop
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

YARN-7374. Improve performance of DRF comparisons for resource types in fair scheduler

This commit is contained in:
Daniel Templeton 2017-10-29 17:45:46 -07:00
parent d4811c8cfa
commit 9711b78998
4 changed files with 306 additions and 42 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/api/records/Resource.java
View File

@ -22,6 +22,7 @@ import java.util.Arrays;
import org.apache.commons.lang.NotImplementedException;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.classification.InterfaceAudience.Private;
import org.apache.hadoop.classification.InterfaceAudience.Public;
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.classification.InterfaceStability.Evolving;
@ -66,8 +67,10 @@ public abstract class Resource implements Comparable<Resource> {
// copy array, etc.
protected static final int NUM_MANDATORY_RESOURCES = 2;
protected static final int MEMORY_INDEX = 0;
protected static final int VCORES_INDEX = 1;
@Private
public static final int MEMORY_INDEX = 0;
@Private
public static final int VCORES_INDEX = 1;
@Public
@Stable

1
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-api/src/main/java/org/apache/hadoop/yarn/api/records/ResourceInformation.java
View File

@ -19,7 +19,6 @@
package org.apache.hadoop.yarn.api.records;
import com.google.common.collect.ImmutableMap;
import org.apache.curator.shaded.com.google.common.reflect.ClassPath;
import org.apache.hadoop.classification.InterfaceAudience;
import org.apache.hadoop.yarn.api.protocolrecords.ResourceTypes;
import org.apache.hadoop.yarn.util.UnitsConversionUtil;

238
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/DominantResourceFairnessPolicy.java
View File

@ -47,8 +47,12 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
public static final String NAME = "DRF";
private static final DominantResourceFairnessComparator COMPARATOR =
new DominantResourceFairnessComparator();
private static final int NUM_RESOURCES =
ResourceUtils.getNumberOfKnownResourceTypes();
private static final DominantResourceFairnessComparator COMPARATORN =
new DominantResourceFairnessComparatorN();
private static final DominantResourceFairnessComparator COMPARATOR2 =
new DominantResourceFairnessComparator2();
private static final DominantResourceCalculator CALCULATOR =
new DominantResourceCalculator();
@ -59,7 +63,15 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
@Override
public Comparator<Schedulable> getComparator() {
return COMPARATOR;
if (NUM_RESOURCES == 2) {
// To improve performance, if we know we're dealing with the common
// case of only CPU and memory, then handle CPU and memory explicitly.
return COMPARATOR2;
} else {
// Otherwise, do it the generic way.
return COMPARATORN;
}
}
@Override
@ -107,25 +119,56 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
@Override
public void initialize(FSContext fsContext) {
COMPARATOR.setFSContext(fsContext);
COMPARATORN.setFSContext(fsContext);
COMPARATOR2.setFSContext(fsContext);
}
/**
* This class compares two {@link Schedulable} instances according to the
* DRF policy. If neither instance is below min share, approximate fair share
* ratios are compared.
* ratios are compared. Subclasses of this class will do the actual work of
* the comparison, specialized for the number of configured resource types.
*/
public static class DominantResourceFairnessComparator
public abstract static class DominantResourceFairnessComparator
implements Comparator<Schedulable> {
private FSContext fsContext;
protected FSContext fsContext;
public void setFSContext(FSContext fsContext) {
this.fsContext = fsContext;
}
/**
* This method is used when apps are tied in fairness ratio. It breaks
* the tie by submit time and job name to get a deterministic ordering,
* which is useful for unit tests.
*
* @param s1 the first item to compare
* @param s2 the second item to compare
* @return &lt; 0, 0, or &gt; 0 if the first item is less than, equal to,
* or greater than the second item, respectively
*/
protected int compareAttribrutes(Schedulable s1, Schedulable s2) {
int res = (int) Math.signum(s1.getStartTime() - s2.getStartTime());
if (res == 0) {
res = s1.getName().compareTo(s2.getName());
}
return res;
}
}
/**
* This class compares two {@link Schedulable} instances according to the
* DRF policy. If neither instance is below min share, approximate fair share
* ratios are compared. This class makes no assumptions about the number of
* resource types.
*/
@VisibleForTesting
static class DominantResourceFairnessComparatorN
extends DominantResourceFairnessComparator {
@Override
public int compare(Schedulable s1, Schedulable s2) {
ResourceInformation[] info = ResourceUtils.getResourceTypesArray();
Resource usage1 = s1.getResourceUsage();
Resource usage2 = s2.getResourceUsage();
Resource minShare1 = s1.getMinShare();
@ -135,8 +178,8 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
// These arrays hold the usage, fair, and min share ratios for each
// resource type. ratios[0][x] are the usage ratios, ratios[1][x] are
// the fair share ratios, and ratios[2][x] are the min share ratios.
float[][] ratios1 = new float[info.length][3];
float[][] ratios2 = new float[info.length][3];
float[][] ratios1 = new float[NUM_RESOURCES][3];
float[][] ratios2 = new float[NUM_RESOURCES][3];
// Calculate cluster shares and approximate fair shares for each
// resource type of both schedulables.
@ -155,7 +198,7 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
usage2.getResources()[dominant2].getValue() <
minShare2.getResources()[dominant2].getValue();
int res = 0;
int res;
if (!s2Needy && !s1Needy) {
// Sort shares by usage ratio and compare them by approximate fair share
@ -176,13 +219,7 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
}
if (res == 0) {
// Apps are tied in fairness ratio. Break the tie by submit time and job
// name to get a deterministic ordering, which is useful for unit tests.
res = (int) Math.signum(s1.getStartTime() - s2.getStartTime());
if (res == 0) {
res = s1.getName().compareTo(s2.getName());
}
res = compareAttribrutes(s1, s2);
}
return res;
@ -206,7 +243,7 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
/**
* Calculate a resource's usage ratio and approximate fair share ratio.
* The {@code shares} array will be populated with both the usage ratio
* The {@code ratios} array will be populated with both the usage ratio
* and the approximate fair share ratio for each resource type. The usage
* ratio is calculated as {@code resource} divided by {@code cluster}.
* The approximate fair share ratio is calculated as the usage ratio
@ -221,18 +258,18 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
* because when comparing resources, the resource with the higher weight
* will be assigned by the scheduler a proportionally higher fair share.
*
* The {@code shares} array must be at least <i>n</i> x 2, where <i>n</i>
* The {@code ratios} array must be at least <i>n</i> x 2, where <i>n</i>
* is the number of resource types. Only the first and second indices of
* the inner arrays in the {@code shares} array will be used, e.g.
* {@code shares[x][0]} and {@code shares[x][1]}.
* the inner arrays in the {@code ratios} array will be used, e.g.
* {@code ratios[x][0]} and {@code ratios[x][1]}.
*
* The return value will be the index of the dominant resource type in the
* {@code shares} array. The dominant resource is the resource type for
* {@code ratios} array. The dominant resource is the resource type for
* which {@code resource} has the largest usage ratio.
*
* @param resource the resource for which to calculate ratios
* @param cluster the total cluster resources
* @param ratios the shares array to populate
* @param ratios the share ratios array to populate
* @param weight the resource weight
* @return the index of the resource type with the largest cluster share
*/
@ -275,7 +312,7 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
*
* @param resource the resource for which to calculate min shares
* @param minShare the min share
* @param ratios the shares array to populate
* @param ratios the share ratios array to populate
*/
@VisibleForTesting
void calculateMinShareRatios(Resource resource, Resource minShare,
@ -320,4 +357,155 @@ public class DominantResourceFairnessPolicy extends SchedulingPolicy {
return ret;
}
}
/**
* This class compares two {@link Schedulable} instances according to the
* DRF policy in the special case that only CPU and memory are configured.
* If neither instance is below min share, approximate fair share
* ratios are compared.
*/
@VisibleForTesting
static class DominantResourceFairnessComparator2
extends DominantResourceFairnessComparator {
@Override
public int compare(Schedulable s1, Schedulable s2) {
ResourceInformation[] resourceInfo1 =
s1.getResourceUsage().getResources();
ResourceInformation[] resourceInfo2 =
s2.getResourceUsage().getResources();
ResourceInformation[] minShareInfo1 = s1.getMinShare().getResources();
ResourceInformation[] minShareInfo2 = s2.getMinShare().getResources();
ResourceInformation[] clusterInfo =
fsContext.getClusterResource().getResources();
double[] shares1 = new double[2];
double[] shares2 = new double[2];
int dominant1 = calculateClusterAndFairRatios(resourceInfo1,
s1.getWeight(), clusterInfo, shares1);
int dominant2 = calculateClusterAndFairRatios(resourceInfo2,
s2.getWeight(), clusterInfo, shares2);
// A queue is needy for its min share if its dominant resource
// (with respect to the cluster capacity) is below its configured min
// share for that resource
boolean s1Needy = resourceInfo1[dominant1].getValue() <
minShareInfo1[dominant1].getValue();
boolean s2Needy = resourceInfo1[dominant2].getValue() <
minShareInfo2[dominant2].getValue();
int res;
if (!s2Needy && !s1Needy) {
res = (int) Math.signum(shares1[dominant1] - shares2[dominant2]);
if (res == 0) {
// Because memory and CPU are indices 0 and 1, we can find the
// non-dominant index by subtracting the dominant index from 1.
res = (int) Math.signum(shares1[1 - dominant1] -
shares2[1 - dominant2]);
}
} else if (s1Needy && !s2Needy) {
res = -1;
} else if (s2Needy && !s1Needy) {
res = 1;
} else {
double[] minShares1 =
calculateMinShareRatios(resourceInfo1, minShareInfo1);
double[] minShares2 =
calculateMinShareRatios(resourceInfo2, minShareInfo2);
res = (int) Math.signum(minShares1[dominant1] - minShares2[dominant2]);
if (res == 0) {
res = (int) Math.signum(minShares1[1 - dominant1] -
minShares2[1 - dominant2]);
}
}
if (res == 0) {
res = compareAttribrutes(s1, s2);
}
return res;
}
/**
* Calculate a resource's usage ratio and approximate fair share ratio
* assuming that CPU and memory are the only configured resource types.
* The {@code shares} array will be populated with the approximate fair
* share ratio for each resource type. The approximate fair share ratio
* is calculated as {@code resourceInfo} divided by {@code cluster} and
* the {@code weight}. If the cluster's resources are 100MB and
* 10 vcores, the usage ({@code resourceInfo}) is 10 MB and 5 CPU, and the
* weights are 2, the fair share ratios will be 0.05 and 0.25.
*
* The approximate fair share ratio is the usage divided by the
* approximate fair share, i.e. the cluster resources times the weight.
* The approximate fair share is an acceptable proxy for the fair share
* because when comparing resources, the resource with the higher weight
* will be assigned by the scheduler a proportionally higher fair share.
*
* The length of the {@code shares} array must be at least 2.
*
* The return value will be the index of the dominant resource type in the
* {@code shares} array. The dominant resource is the resource type for
* which {@code resourceInfo} has the largest usage ratio.
*
* @param resourceInfo the resource for which to calculate ratios
* @param weight the resource weight
* @param clusterInfo the total cluster resources
* @param shares the share ratios array to populate
* @return the index of the resource type with the largest cluster share
*/
@VisibleForTesting
int calculateClusterAndFairRatios(ResourceInformation[] resourceInfo,
float weight, ResourceInformation[] clusterInfo, double[] shares) {
int dominant;
shares[Resource.MEMORY_INDEX] =
((double) resourceInfo[Resource.MEMORY_INDEX].getValue()) /
clusterInfo[Resource.MEMORY_INDEX].getValue();
shares[Resource.VCORES_INDEX] =
((double) resourceInfo[Resource.VCORES_INDEX].getValue()) /
clusterInfo[Resource.VCORES_INDEX].getValue();
dominant =
shares[Resource.VCORES_INDEX] > shares[Resource.MEMORY_INDEX] ?
Resource.VCORES_INDEX : Resource.MEMORY_INDEX;
shares[Resource.MEMORY_INDEX] /= weight;
shares[Resource.VCORES_INDEX] /= weight;
return dominant;
}
/**
* Calculate a resource's min share ratios assuming that CPU and memory
* are the only configured resource types. The return array will be
* populated with the {@code resourceInfo} divided by {@code minShareInfo}
* for each resource type. If the min shares are 5 MB and 10 vcores, and
* the usage ({@code resourceInfo}) is 10 MB and 5 CPU, the ratios will
* be 2 and 0.5.
*
* The length of the {@code ratios} array must be 2.
*
* @param resourceInfo the resource for which to calculate min shares
* @param minShareInfo the min share
* @return the share ratios
*/
@VisibleForTesting
double[] calculateMinShareRatios(ResourceInformation[] resourceInfo,
ResourceInformation[] minShareInfo) {
double[] minShares1 = new double[2];
// both are needy below min share
minShares1[Resource.MEMORY_INDEX] =
((double) resourceInfo[Resource.MEMORY_INDEX].getValue()) /
minShareInfo[Resource.MEMORY_INDEX].getValue();
minShares1[Resource.VCORES_INDEX] =
((double) resourceInfo[Resource.VCORES_INDEX].getValue()) /
minShareInfo[Resource.VCORES_INDEX].getValue();
return minShares1;
}
}
}

102
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/test/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/fair/policies/TestDominantResourceFairnessPolicy.java
View File

@ -24,21 +24,22 @@ import static org.mockito.Mockito.when;
import java.util.Comparator;
import java.util.Map;
import org.apache.curator.shaded.com.google.common.base.Joiner;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.yarn.api.records.Resource;
import org.apache.hadoop.yarn.api.records.ResourceInformation;
import org.apache.hadoop.yarn.conf.YarnConfiguration;
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FSContext;
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FakeSchedulable;
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.Schedulable;
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.policies.DominantResourceFairnessPolicy.DominantResourceFairnessComparator;
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.policies.DominantResourceFairnessPolicy.DominantResourceFairnessComparatorN;
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.policies.DominantResourceFairnessPolicy.DominantResourceFairnessComparator2;
import org.apache.hadoop.yarn.server.utils.BuilderUtils;
import org.apache.hadoop.yarn.util.resource.ResourceUtils;
import org.apache.hadoop.yarn.util.resource.Resources;
import org.junit.Assert;
import org.junit.BeforeClass;
import org.junit.Before;
import org.junit.Test;
/**
@ -46,8 +47,8 @@ import org.junit.Test;
* container before sched2
*/
public class TestDominantResourceFairnessPolicy {
@BeforeClass
public static void setup() {
@Before
public void setup() {
addResources("test");
}
@ -77,7 +78,6 @@ public class TestDominantResourceFairnessPolicy {
return createSchedulable(memUsage, cpuUsage, weights, 0, 0);
}
private Schedulable createSchedulable(int memUsage, int cpuUsage,
float weights, int minMemShare, int minCpuShare) {
Resource usage = BuilderUtils.newResource(memUsage, cpuUsage);
@ -97,6 +97,12 @@ public class TestDominantResourceFairnessPolicy {
c.compare(s1, s2) < 0);
}
@Test
public void testSameDominantResource2() {
ResourceUtils.resetResourceTypes(new Configuration());
testSameDominantResource();
}
@Test
public void testDifferentDominantResource() {
Comparator c = createComparator(8000, 8);
@ -107,6 +113,12 @@ public class TestDominantResourceFairnessPolicy {
c.compare(s1, s2) < 0);
}
@Test
public void testDifferentDominantResource2() {
ResourceUtils.resetResourceTypes(new Configuration());
testDifferentDominantResource();
}
@Test
public void testOneIsNeedy() {
Comparator c = createComparator(8000, 8);
@ -117,6 +129,12 @@ public class TestDominantResourceFairnessPolicy {
c.compare(s1, s2) < 0);
}
@Test
public void testOneIsNeedy2() {
ResourceUtils.resetResourceTypes(new Configuration());
testOneIsNeedy();
}
@Test
public void testBothAreNeedy() {
Comparator c = createComparator(8000, 100);
@ -137,6 +155,12 @@ public class TestDominantResourceFairnessPolicy {
c.compare(s1, s2) < 0);
}
@Test
public void testBothAreNeedy2() {
ResourceUtils.resetResourceTypes(new Configuration());
testBothAreNeedy();
}
@Test
public void testEvenWeightsSameDominantResource() {
assertTrue(createComparator(8000, 8).compare(
@ -147,6 +171,12 @@ public class TestDominantResourceFairnessPolicy {
createSchedulable(1000, 2)) < 0);
}
@Test
public void testEvenWeightsSameDominantResource2() {
ResourceUtils.resetResourceTypes(new Configuration());
testEvenWeightsSameDominantResource();
}
@Test
public void testEvenWeightsDifferentDominantResource() {
assertTrue(createComparator(8000, 8).compare(
@ -157,14 +187,20 @@ public class TestDominantResourceFairnessPolicy {
createSchedulable(1000, 2)) < 0);
}
@Test
public void testEvenWeightsDifferentDominantResource2() {
ResourceUtils.resetResourceTypes(new Configuration());
testEvenWeightsDifferentDominantResource();
}
@Test
public void testSortShares() {
float[][] ratios1 = {{0.3f, 2.0f}, {0.2f, 1.0f}, {0.4f, 0.1f}};
float[][] ratios2 = {{0.2f, 9.0f}, {0.3f, 2.0f}, {0.25f, 0.1f}};
float[][] expected1 = {{0.4f, 0.1f}, {0.3f, 2.0f}, {0.2f, 1.0f}};
float[][] expected2 = {{0.3f, 2.0f}, {0.25f, 0.1f}, {0.2f, 9.0f}};
DominantResourceFairnessComparator comparator =
new DominantResourceFairnessComparator();
DominantResourceFairnessComparatorN comparator =
new DominantResourceFairnessComparatorN();
comparator.sortRatios(ratios1, ratios2);
@ -184,8 +220,8 @@ public class TestDominantResourceFairnessPolicy {
Resource used = Resources.createResource(10, 5);
Resource capacity = Resources.createResource(100, 10);
float[][] shares = new float[3][2];
DominantResourceFairnessComparator comparator =
new DominantResourceFairnessComparator();
DominantResourceFairnessComparatorN comparator =
new DominantResourceFairnessComparatorN();
used.setResourceValue("test", 2L);
capacity.setResourceValue("test", 5L);
@ -206,14 +242,34 @@ public class TestDominantResourceFairnessPolicy {
dominant);
}
@Test
public void testCalculateClusterAndFairRatios2() {
ResourceUtils.resetResourceTypes(new Configuration());
Resource used = Resources.createResource(10, 5);
Resource capacity = Resources.createResource(100, 10);
double[] shares = new double[2];
DominantResourceFairnessComparator2 comparator =
new DominantResourceFairnessComparator2();
int dominant =
comparator.calculateClusterAndFairRatios(used.getResources(), 1.0f,
capacity.getResources(), shares);
assertEquals("Calculated usage ratio for memory (10MB out of 100MB) is "
+ "incorrect", 0.1, shares[Resource.MEMORY_INDEX], .00001);
assertEquals("Calculated usage ratio for vcores (5 out of 10) is "
+ "incorrect", 0.5, shares[Resource.VCORES_INDEX], .00001);
assertEquals("The wrong dominant resource index was returned",
Resource.VCORES_INDEX, dominant);
}
@Test
public void testCalculateMinShareRatios() {
Map<String, Integer> index = ResourceUtils.getResourceTypeIndex();
Resource used = Resources.createResource(10, 5);
Resource minShares = Resources.createResource(5, 10);
float[][] ratios = new float[3][3];
DominantResourceFairnessComparator comparator =
new DominantResourceFairnessComparator();
DominantResourceFairnessComparatorN comparator =
new DominantResourceFairnessComparatorN();
used.setResourceValue("test", 2L);
minShares.setResourceValue("test", 0L);
@ -231,6 +287,24 @@ public class TestDominantResourceFairnessPolicy {
0.00001f);
}
@Test
public void testCalculateMinShareRatios2() {
ResourceUtils.resetResourceTypes(new Configuration());
Resource used = Resources.createResource(10, 5);
Resource minShares = Resources.createResource(5, 10);
DominantResourceFairnessComparator2 comparator =
new DominantResourceFairnessComparator2();
double[] ratios =
comparator.calculateMinShareRatios(used.getResources(),
minShares.getResources());
assertEquals("Calculated min share ratio for memory (10MB out of 5MB) is "
+ "incorrect", 2.0, ratios[Resource.MEMORY_INDEX], .00001f);
assertEquals("Calculated min share ratio for vcores (5 out of 10) is "
+ "incorrect", 0.5, ratios[Resource.VCORES_INDEX], .00001f);
}
@Test
public void testCompareShares() {
float[][] ratios1 = {
@ -248,8 +322,8 @@ public class TestDominantResourceFairnessPolicy {
{0.2f, 0.1f, 2.0f},
{0.1f, 2.0f, 1.0f}
};
DominantResourceFairnessComparator comparator =
new DominantResourceFairnessComparator();
DominantResourceFairnessComparatorN comparator =
new DominantResourceFairnessComparatorN();
int ret = comparator.compareRatios(ratios1, ratios2, 0);