Skip to content

HBASE-24138 log more details about balancer decisions for StochasticLoadBalancer #1455

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Apr 8, 2020
Merged

HBASE-24138 log more details about balancer decisions for StochasticLoadBalancer #1455

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0786490
HBASE-24138 log more details about balancer decisions for StochasticL…
busbey Apr 3, 2020
e82b288
HBASE-24138 checkstyle fix per qabot.
busbey Apr 8, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions hbase-server/src/main/java/org/apache/hadoop/hbase/master/balancer/ServerAndLoad.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,4 +65,9 @@ public boolean equals(Object o) {
}
return false;
}

@Override
public String toString() {
return "server=" + sn + " , load=" + load;
}
}
36 changes: 24 additions & 12 deletions ...-server/src/main/java/org/apache/hadoop/hbase/master/balancer/StochasticLoadBalancer.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -331,28 +331,29 @@ protected boolean needsBalance(TableName tableName, Cluster cluster) {
for (CostFunction c : costFunctions) {
float multiplier = c.getMultiplier();
if (multiplier <= 0) {
LOG.trace("{} not needed because multiplier is <= 0", c.getClass().getSimpleName());
continue;
}
if (!c.isNeeded()) {
LOG.debug("{} not needed", c.getClass().getSimpleName());
LOG.trace("{} not needed", c.getClass().getSimpleName());
continue;
}
sumMultiplier += multiplier;
total += c.cost() * multiplier;
}

if (total <= 0 || sumMultiplier <= 0
|| (sumMultiplier > 0 && (total / sumMultiplier) < minCostNeedBalance)) {
boolean balanced = total <= 0 || sumMultiplier <= 0 ||
(sumMultiplier > 0 && (total / sumMultiplier) < minCostNeedBalance);
if (LOG.isDebugEnabled()) {
LOG.debug("{} {}; total cost={}, sum multiplier={}; cost/multiplier to need a balance is {}",
balanced ? "Skipping load balancing because balanced" : "We need to load balance",
isByTable ? String.format("table (%s)", tableName) : "cluster",
total, sumMultiplier, minCostNeedBalance);
if (LOG.isTraceEnabled()) {
final String loadBalanceTarget =
isByTable ? String.format("table (%s)", tableName) : "cluster";
LOG.trace("Skipping load balancing because the {} is balanced. Total cost: {}, "
+ "Sum multiplier: {}, Minimum cost needed for balance: {}", loadBalanceTarget, total,
sumMultiplier, minCostNeedBalance);
LOG.trace("Balance decision detailed function costs={}", functionCost());
}
return false;
}
return true;
return !balanced;
}

@VisibleForTesting
Expand Down Expand Up @@ -1189,16 +1190,27 @@ static class RegionCountSkewCostFunction extends CostFunction {
this.setMultiplier(conf.getFloat(REGION_COUNT_SKEW_COST_KEY, DEFAULT_REGION_COUNT_SKEW_COST));
}

@Override
void init(Cluster cluster) {
super.init(cluster);
LOG.debug("{} sees a total of {} servers and {} regions.", getClass().getSimpleName(),
cluster.numServers, cluster.numRegions);
if (LOG.isTraceEnabled()) {
for (int i =0; i < cluster.numServers; i++) {
LOG.trace("{} sees server '{}' has {} regions", getClass().getSimpleName(),
cluster.servers[i], cluster.regionsPerServer[i].length);
}
}
}

@Override
protected double cost() {
if (stats == null || stats.length != cluster.numServers) {
stats = new double[cluster.numServers];
}

for (int i =0; i < cluster.numServers; i++) {
stats[i] = cluster.regionsPerServer[i].length;
}

return costFromArray(stats);
}
}
Expand Down
13 changes: 8 additions & 5 deletions hbase-server/src/test/java/org/apache/hadoop/hbase/master/balancer/BalancerTestBase.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -201,9 +201,11 @@ public void assertClusterAsBalanced(List<ServerAndLoad> servers) {
int max = numRegions % numServers == 0 ? min : min + 1;

for (ServerAndLoad server : servers) {
assertTrue(server.getLoad() >= 0);
assertTrue(server.getLoad() <= max);
assertTrue(server.getLoad() >= min);
assertTrue("All servers should have a positive load. " + server, server.getLoad() >= 0);
assertTrue("All servers should have load no more than " + max + ". " + server,
server.getLoad() <= max);
assertTrue("All servers should have load no less than " + min + ". " + server,
server.getLoad() >= min);
}
}

Expand Down Expand Up @@ -561,7 +563,7 @@ protected void testWithCluster(Map<ServerName, List<RegionInfo>> serverMap,
Map<TableName, Map<ServerName, List<RegionInfo>>> LoadOfAllTable =
(Map) mockClusterServersWithTables(serverMap);
List<RegionPlan> plans = loadBalancer.balanceCluster(LoadOfAllTable);
assertNotNull(plans);
assertNotNull("Initial cluster balance should produce plans.", plans);

// Check to see that this actually got to a stable place.
if (assertFullyBalanced || assertFullyBalancedForReplicas) {
Expand All @@ -575,7 +577,8 @@ protected void testWithCluster(Map<ServerName, List<RegionInfo>> serverMap,
assertClusterAsBalanced(balancedCluster);
LoadOfAllTable = (Map) mockClusterServersWithTables(serverMap);
List<RegionPlan> secondPlans = loadBalancer.balanceCluster(LoadOfAllTable);
assertNull(secondPlans);
assertNull("Given a requirement to be fully balanced, second attempt at plans should " +
"produce none.", secondPlans);
}

if (assertFullyBalancedForReplicas) {
Expand Down