001/* 002 * Licensed to the Apache Software Foundation (ASF) under one 003 * or more contributor license agreements. See the NOTICE file 004 * distributed with this work for additional information 005 * regarding copyright ownership. The ASF licenses this file 006 * to you under the Apache License, Version 2.0 (the 007 * "License"); you may not use this file except in compliance 008 * with the License. You may obtain a copy of the License at 009 * 010 * http://www.apache.org/licenses/LICENSE-2.0 011 * 012 * Unless required by applicable law or agreed to in writing, software 013 * distributed under the License is distributed on an "AS IS" BASIS, 014 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 015 * See the License for the specific language governing permissions and 016 * limitations under the License. 017 */ 018package org.apache.hadoop.hbase.master.balancer; 019 020import java.util.concurrent.atomic.AtomicLong; 021import org.agrona.collections.Hashing; 022import org.agrona.collections.Int2IntCounterMap; 023import org.apache.yetus.audience.InterfaceAudience; 024 025/** 026 * A cost function for region replicas. We give a high cost for hosting replicas of the same region 027 * in the same server, host or rack. We do not prevent the case though, since if numReplicas > 028 * numRegionServers, we still want to keep the replica open. 029 */ 030@InterfaceAudience.Private 031abstract class RegionReplicaGroupingCostFunction extends CostFunction { 032 protected long maxCost = 0; 033 protected long[] costsPerGroup; // group is either server, host or rack 034 035 @Override 036 final void prepare(BalancerClusterState cluster) { 037 super.prepare(cluster); 038 if (!isNeeded()) { 039 return; 040 } 041 loadCosts(); 042 } 043 044 protected abstract void loadCosts(); 045 046 protected final long getMaxCost(BalancerClusterState cluster) { 047 // max cost is the case where every region replica is hosted together regardless of host 048 Int2IntCounterMap colocatedReplicaCounts = 049 new Int2IntCounterMap(cluster.numRegions, Hashing.DEFAULT_LOAD_FACTOR, 0); 050 for (int i = 0; i < cluster.regionIndexToPrimaryIndex.length; i++) { 051 colocatedReplicaCounts.getAndIncrement(cluster.regionIndexToPrimaryIndex[i]); 052 } 053 // compute numReplicas from the sorted array 054 return costPerGroup(colocatedReplicaCounts); 055 } 056 057 @Override 058 boolean isNeeded() { 059 return cluster.hasRegionReplicas; 060 } 061 062 @Override 063 protected double cost() { 064 if (maxCost <= 0) { 065 return 0; 066 } 067 068 long totalCost = 0; 069 for (int i = 0; i < costsPerGroup.length; i++) { 070 totalCost += costsPerGroup[i]; 071 } 072 return scale(0, maxCost, totalCost); 073 } 074 075 /** 076 * For each primary region, it computes the total number of replicas in the array (numReplicas) 077 * and returns a sum of numReplicas-1 squared. For example, if the server hosts regions a, b, c, 078 * d, e, f where a and b are same replicas, and c,d,e are same replicas, it returns (2-1) * (2-1) 079 * + (3-1) * (3-1) + (1-1) * (1-1). 080 * @param colocatedReplicaCounts a sorted array of primary regions ids for the regions hosted 081 * @return a sum of numReplicas-1 squared for each primary region in the group. 082 */ 083 protected final long costPerGroup(Int2IntCounterMap colocatedReplicaCounts) { 084 final AtomicLong cost = new AtomicLong(0); 085 // colocatedReplicaCounts is a sorted array of primary ids of regions. Replicas of regions 086 // sharing the same primary will have consecutive numbers in the array. 087 colocatedReplicaCounts.forEach((primary, count) -> { 088 if (count > 1) { // means consecutive primaries, indicating co-location 089 cost.getAndAdd((count - 1) * (count - 1)); 090 } 091 }); 092 return cost.longValue(); 093 } 094 095 @Override 096 public final void updateWeight(double[] weights) { 097 weights[StochasticLoadBalancer.GeneratorType.RACK.ordinal()] += cost(); 098 } 099}