cassandra-analytics-core/src/main/java/org/apache/cassandra/spark/bulkwriter/token/ConsistencyLevel.java - cassandra-analytics - Git at Google

 /*
  * Licensed to the Apache Software Foundation (ASF) under one
  * or more contributor license agreements.  See the NOTICE file
  * distributed with this work for additional information
  * regarding copyright ownership.  The ASF licenses this file
  * to you under the Apache License, Version 2.0 (the
  * "License"); you may not use this file except in compliance
  * with the License.  You may obtain a copy of the License at
  *
  *   http://www.apache.org/licenses/LICENSE-2.0
  *
  * Unless required by applicable law or agreed to in writing,
  * software distributed under the License is distributed on an
  * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
  * KIND, either express or implied.  See the License for the
  * specific language governing permissions and limitations
  * under the License.
  */

 package org.apache.cassandra.spark.bulkwriter.token;

 import java.util.Collection;
 import java.util.Objects;
 import java.util.Set;

 import com.google.common.base.Preconditions;

 import org.apache.cassandra.spark.common.model.CassandraInstance;
 import org.apache.cassandra.spark.data.ReplicationFactor;

 public interface ConsistencyLevel
 {
     /**
      * Whether the consistency level only considers replicas in the local data center.
      *
      * @return true if only considering the local replicas; otherwise, return false
      */
     boolean isLocal();

     /**
      * Check consistency level with the collection of the succeeded instances
      *
      * @param succeededInstances the succeeded instances in the replica set
      * @param replicationFactor replication factor to check with
      * @param localDC the local data center name if required for the check
      * @return true means the consistency level is _definitively_ satisfied.
      *         Meanwhile, returning false means no conclusion can be drawn
      */
     boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
                            ReplicationFactor replicationFactor,
                            String localDC);

     default void ensureNetworkTopologyStrategy(ReplicationFactor replicationFactor, CL cl)
     {
         Preconditions.checkArgument(replicationFactor.getReplicationStrategy() == ReplicationFactor.ReplicationStrategy.NetworkTopologyStrategy,
                                     cl.name() + " only make sense for NetworkTopologyStrategy keyspaces");
     }

     /**
      * Checks if the consistency guarantees are maintained, given the failed, blocked and replacing instances, consistency-level and the replication-factor.
      * <pre>
      * - QUORUM based consistency levels check for quorum using the write-replica-set (instead of RF) as they include healthy and pending nodes.
      *   This is done to ensure that writes go to a quorum of healthy nodes while accounting for potential failure in pending nodes becoming healthy.
      * - ONE and TWO consistency guarantees are maintained by ensuring that the failures leave us with at-least the corresponding healthy
      *   (and non-pending) nodes.
      *
      *   For both the above cases, blocked instances are also considered as failures while performing consistency checks.
      *   Write replicas are adjusted to exclude replacement nodes for consistency checks, if we have replacement nodes that are not among the failed instances.
      *   This is to ensure that we are writing to sufficient non-replacement nodes as replacements can potentially fail leaving us with fewer nodes.
      * </pre>
      *
      * @param writeReplicas        the set of replicas for write operations
      * @param pendingReplicas      the set of replicas pending status
      * @param replacementInstances the set of instances that are replacing the other instances
      * @param blockedInstances     the set of instances that have been blocked for the bulk operation
      * @param failedInstanceIps    the set of instances where there were failures
      * @param localDC              the local datacenter used for consistency level, or {@code null} if not provided
      * @return {@code true} if the consistency has been met, {@code false} otherwise
      */
     boolean checkConsistency(Set<String> writeReplicas,
                              Set<String> pendingReplicas,
                              Set<String> replacementInstances,
                              Set<String> blockedInstances,
                              Set<String> failedInstanceIps,
                              String localDC); // todo: simplify the parameter list. not all are required in impl

     // Check if successful writes forms quorum of non-replacing nodes - N/A as quorum is if there are no failures/blocked
     enum CL implements ConsistencyLevel
     {
         ALL
         {
             @Override
             public boolean isLocal()
             {
                 return false;
             }

             @Override
             public boolean checkConsistency(Set<String> writeReplicas,
                                             Set<String> pendingReplicas,
                                             Set<String> replacementInstances,
                                             Set<String> blockedInstances,
                                             Set<String> failedInstanceIps,
                                             String localDC)
             {
                 int failedExcludingReplacements = failedInstanceIps.size() - replacementInstances.size();
                 return failedExcludingReplacements <= 0 && blockedInstances.isEmpty();
             }

             @Override
             public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
                                           ReplicationFactor replicationFactor,
                                           String localDC)
             {
                 int rf = replicationFactor.getTotalReplicationFactor();
                 // The effective RF during expansion could be larger than the defined RF
                 // The check for CL satisfaction should consider the scenario and use >=
                 return succeededInstances.size() >= rf;
             }
         },

         EACH_QUORUM
         {
             @Override
             public boolean isLocal()
             {
                 return false;
             }

             @Override
             public boolean checkConsistency(Set<String> writeReplicas,
                                             Set<String> pendingReplicas,
                                             Set<String> replacementInstances,
                                             Set<String> blockedInstances,
                                             Set<String> failedInstanceIps,
                                             String localDC)
             {
                 return (failedInstanceIps.size() + blockedInstances.size()) <= (writeReplicas.size() - (writeReplicas.size() / 2 + 1));
             }

             @Override
             public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
                                           ReplicationFactor replicationFactor,
                                           String localDC)
             {
                 ensureNetworkTopologyStrategy(replicationFactor, EACH_QUORUM);
                 Objects.requireNonNull(localDC, "localDC cannot be null");

                 for (String datacenter : replicationFactor.getOptions().keySet())
                 {
                     int rf = replicationFactor.getOptions().get(datacenter);
                     int majority = rf / 2 + 1;
                     if (succeededInstances.stream()
                                           .filter(instance -> instance.datacenter().equalsIgnoreCase(datacenter))
                                           .count() < majority)
                     {
                         return false;
                     }
                 }
                 return true;
             }
         },
         QUORUM
         {
             @Override
             public boolean isLocal()
             {
                 return false;
             }

             @Override
             public boolean checkConsistency(Set<String> writeReplicas,
                                             Set<String> pendingReplicas,
                                             Set<String> replacementInstances,
                                             Set<String> blockedInstances,
                                             Set<String> failedInstanceIps,
                                             String localDC)
             {
                 return (failedInstanceIps.size() + blockedInstances.size()) <= (writeReplicas.size() - (writeReplicas.size() / 2 + 1));
             }

             @Override
             public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
                                           ReplicationFactor replicationFactor,
                                           String localDC)
             {
                 int rf = replicationFactor.getTotalReplicationFactor();
                 return succeededInstances.size() > rf / 2;
             }
         },
         LOCAL_QUORUM
         {
             @Override
             public boolean isLocal()
             {
                 return true;
             }

             @Override
             public boolean checkConsistency(Set<String> writeReplicas,
                                             Set<String> pendingReplicas,
                                             Set<String> replacementInstances,
                                             Set<String> blockedInstances,
                                             Set<String> failedInstanceIps,
                                             String localDC)
             {
                 return (failedInstanceIps.size() + blockedInstances.size()) <= (writeReplicas.size() - (writeReplicas.size() / 2 + 1));
             }

             @Override
             public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
                                           ReplicationFactor replicationFactor,
                                           String localDC)
             {
                 ensureNetworkTopologyStrategy(replicationFactor, LOCAL_QUORUM);
                 Objects.requireNonNull(localDC, "localDC cannot be null");

                 int rf = replicationFactor.getOptions().get(localDC);
                 return succeededInstances.stream()
                                          .filter(instance -> instance.datacenter().equalsIgnoreCase(localDC))
                                          .count() > rf / 2;
             }
         },
         ONE
         {
             @Override
             public boolean isLocal()
             {
                 return false;
             }

             @Override
             public boolean checkConsistency(Set<String> writeReplicas,
                                             Set<String> pendingReplicas,
                                             Set<String> replacementInstances,
                                             Set<String> blockedInstances,
                                             Set<String> failedInstanceIps,
                                             String localDC)
             {
                 return (failedInstanceIps.size() + blockedInstances.size())
                        <= (writeReplicas.size() - pendingReplicas.size() - 1);
             }

             @Override
             public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
                                           ReplicationFactor replicationFactor,
                                           String localDC)
             {
                 return !succeededInstances.isEmpty();
             }
         },
         TWO
         {
             @Override
             public boolean isLocal()
             {
                 return false;
             }

             @Override
             public boolean checkConsistency(Set<String> writeReplicas,
                                             Set<String> pendingReplicas,
                                             Set<String> replacementInstances,
                                             Set<String> blockedInstances,
                                             Set<String> failedInstanceIps,
                                             String localDC)
             {
                 return (failedInstanceIps.size() + blockedInstances.size())
                        <= (writeReplicas.size() - pendingReplicas.size() - 2);
             }

             @Override
             public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
                                           ReplicationFactor replicationFactor,
                                           String localDC)
             {
                 return succeededInstances.size() >= 2;
             }
         },
         LOCAL_ONE
         {
             @Override
             public boolean isLocal()
             {
                 return true;
             }

             @Override
             public boolean checkConsistency(Set<String> writeReplicas,
                                             Set<String> pendingReplicas,
                                             Set<String> replacementInstances,
                                             Set<String> blockedInstances,
                                             Set<String> failedInstanceIps,
                                             String localDC)
             {
                 return (failedInstanceIps.size() + blockedInstances.size()) <= (writeReplicas.size() - pendingReplicas.size() - 1);
             }

             @Override
             public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
                                           ReplicationFactor replicationFactor,
                                           String localDC)
             {
                 ensureNetworkTopologyStrategy(replicationFactor, LOCAL_ONE);
                 Objects.requireNonNull(localDC, "localDC cannot be null");

                 return succeededInstances.stream().anyMatch(instance -> instance.datacenter().equalsIgnoreCase(localDC));
             }
         };
     }
 }
	/*
	* Licensed to the Apache Software Foundation (ASF) under one
	* or more contributor license agreements. See the NOTICE file
	* distributed with this work for additional information
	* regarding copyright ownership. The ASF licenses this file
	* to you under the Apache License, Version 2.0 (the
	* "License"); you may not use this file except in compliance
	* with the License. You may obtain a copy of the License at
	*
	* http://www.apache.org/licenses/LICENSE-2.0
	*
	* Unless required by applicable law or agreed to in writing,
	* software distributed under the License is distributed on an
	* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
	* KIND, either express or implied. See the License for the
	* specific language governing permissions and limitations
	* under the License.
	*/

	package org.apache.cassandra.spark.bulkwriter.token;

	import java.util.Collection;
	import java.util.Objects;
	import java.util.Set;

	import com.google.common.base.Preconditions;

	import org.apache.cassandra.spark.common.model.CassandraInstance;
	import org.apache.cassandra.spark.data.ReplicationFactor;

	public interface ConsistencyLevel
	{
	/**
	* Whether the consistency level only considers replicas in the local data center.
	*
	* @return true if only considering the local replicas; otherwise, return false
	*/
	boolean isLocal();

	/**
	* Check consistency level with the collection of the succeeded instances
	*
	* @param succeededInstances the succeeded instances in the replica set
	* @param replicationFactor replication factor to check with
	* @param localDC the local data center name if required for the check
	* @return true means the consistency level is _definitively_ satisfied.
	* Meanwhile, returning false means no conclusion can be drawn
	*/
	boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
	ReplicationFactor replicationFactor,
	String localDC);

	default void ensureNetworkTopologyStrategy(ReplicationFactor replicationFactor, CL cl)
	{
	Preconditions.checkArgument(replicationFactor.getReplicationStrategy() == ReplicationFactor.ReplicationStrategy.NetworkTopologyStrategy,
	cl.name() + " only make sense for NetworkTopologyStrategy keyspaces");
	}

	/**
	* Checks if the consistency guarantees are maintained, given the failed, blocked and replacing instances, consistency-level and the replication-factor.
	* <pre>
	* - QUORUM based consistency levels check for quorum using the write-replica-set (instead of RF) as they include healthy and pending nodes.
	* This is done to ensure that writes go to a quorum of healthy nodes while accounting for potential failure in pending nodes becoming healthy.
	* - ONE and TWO consistency guarantees are maintained by ensuring that the failures leave us with at-least the corresponding healthy
	* (and non-pending) nodes.
	*
	* For both the above cases, blocked instances are also considered as failures while performing consistency checks.
	* Write replicas are adjusted to exclude replacement nodes for consistency checks, if we have replacement nodes that are not among the failed instances.
	* This is to ensure that we are writing to sufficient non-replacement nodes as replacements can potentially fail leaving us with fewer nodes.
	* </pre>
	*
	* @param writeReplicas the set of replicas for write operations
	* @param pendingReplicas the set of replicas pending status
	* @param replacementInstances the set of instances that are replacing the other instances
	* @param blockedInstances the set of instances that have been blocked for the bulk operation
	* @param failedInstanceIps the set of instances where there were failures
	* @param localDC the local datacenter used for consistency level, or {@code null} if not provided
	* @return {@code true} if the consistency has been met, {@code false} otherwise
	*/
	boolean checkConsistency(Set<String> writeReplicas,
	Set<String> pendingReplicas,
	Set<String> replacementInstances,
	Set<String> blockedInstances,
	Set<String> failedInstanceIps,
	String localDC); // todo: simplify the parameter list. not all are required in impl

	// Check if successful writes forms quorum of non-replacing nodes - N/A as quorum is if there are no failures/blocked
	enum CL implements ConsistencyLevel
	{
	ALL
	{
	@Override
	public boolean isLocal()
	{
	return false;
	}

	@Override
	public boolean checkConsistency(Set<String> writeReplicas,
	Set<String> pendingReplicas,
	Set<String> replacementInstances,
	Set<String> blockedInstances,
	Set<String> failedInstanceIps,
	String localDC)
	{
	int failedExcludingReplacements = failedInstanceIps.size() - replacementInstances.size();
	return failedExcludingReplacements <= 0 && blockedInstances.isEmpty();
	}

	@Override
	public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
	ReplicationFactor replicationFactor,
	String localDC)
	{
	int rf = replicationFactor.getTotalReplicationFactor();
	// The effective RF during expansion could be larger than the defined RF
	// The check for CL satisfaction should consider the scenario and use >=
	return succeededInstances.size() >= rf;
	}
	},

	EACH_QUORUM
	{
	@Override
	public boolean isLocal()
	{
	return false;
	}

	@Override
	public boolean checkConsistency(Set<String> writeReplicas,
	Set<String> pendingReplicas,
	Set<String> replacementInstances,
	Set<String> blockedInstances,
	Set<String> failedInstanceIps,
	String localDC)
	{
	return (failedInstanceIps.size() + blockedInstances.size()) <= (writeReplicas.size() - (writeReplicas.size() / 2 + 1));
	}

	@Override
	public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
	ReplicationFactor replicationFactor,
	String localDC)
	{
	ensureNetworkTopologyStrategy(replicationFactor, EACH_QUORUM);
	Objects.requireNonNull(localDC, "localDC cannot be null");

	for (String datacenter : replicationFactor.getOptions().keySet())
	{
	int rf = replicationFactor.getOptions().get(datacenter);
	int majority = rf / 2 + 1;
	if (succeededInstances.stream()
	.filter(instance -> instance.datacenter().equalsIgnoreCase(datacenter))
	.count() < majority)
	{
	return false;
	}
	}
	return true;
	}
	},
	QUORUM
	{
	@Override
	public boolean isLocal()
	{
	return false;
	}

	@Override
	public boolean checkConsistency(Set<String> writeReplicas,
	Set<String> pendingReplicas,
	Set<String> replacementInstances,
	Set<String> blockedInstances,
	Set<String> failedInstanceIps,
	String localDC)
	{
	return (failedInstanceIps.size() + blockedInstances.size()) <= (writeReplicas.size() - (writeReplicas.size() / 2 + 1));
	}

	@Override
	public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
	ReplicationFactor replicationFactor,
	String localDC)
	{
	int rf = replicationFactor.getTotalReplicationFactor();
	return succeededInstances.size() > rf / 2;
	}
	},
	LOCAL_QUORUM
	{
	@Override
	public boolean isLocal()
	{
	return true;
	}

	@Override
	public boolean checkConsistency(Set<String> writeReplicas,
	Set<String> pendingReplicas,
	Set<String> replacementInstances,
	Set<String> blockedInstances,
	Set<String> failedInstanceIps,
	String localDC)
	{
	return (failedInstanceIps.size() + blockedInstances.size()) <= (writeReplicas.size() - (writeReplicas.size() / 2 + 1));
	}

	@Override
	public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
	ReplicationFactor replicationFactor,
	String localDC)
	{
	ensureNetworkTopologyStrategy(replicationFactor, LOCAL_QUORUM);
	Objects.requireNonNull(localDC, "localDC cannot be null");

	int rf = replicationFactor.getOptions().get(localDC);
	return succeededInstances.stream()
	.filter(instance -> instance.datacenter().equalsIgnoreCase(localDC))
	.count() > rf / 2;
	}
	},
	ONE
	{
	@Override
	public boolean isLocal()
	{
	return false;
	}

	@Override
	public boolean checkConsistency(Set<String> writeReplicas,
	Set<String> pendingReplicas,
	Set<String> replacementInstances,
	Set<String> blockedInstances,
	Set<String> failedInstanceIps,
	String localDC)
	{
	return (failedInstanceIps.size() + blockedInstances.size())
	<= (writeReplicas.size() - pendingReplicas.size() - 1);
	}

	@Override
	public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
	ReplicationFactor replicationFactor,
	String localDC)
	{
	return !succeededInstances.isEmpty();
	}
	},
	TWO
	{
	@Override
	public boolean isLocal()
	{
	return false;
	}

	@Override
	public boolean checkConsistency(Set<String> writeReplicas,
	Set<String> pendingReplicas,
	Set<String> replacementInstances,
	Set<String> blockedInstances,
	Set<String> failedInstanceIps,
	String localDC)
	{
	return (failedInstanceIps.size() + blockedInstances.size())
	<= (writeReplicas.size() - pendingReplicas.size() - 2);
	}

	@Override
	public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
	ReplicationFactor replicationFactor,
	String localDC)
	{
	return succeededInstances.size() >= 2;
	}
	},
	LOCAL_ONE
	{
	@Override
	public boolean isLocal()
	{
	return true;
	}

	@Override
	public boolean checkConsistency(Set<String> writeReplicas,
	Set<String> pendingReplicas,
	Set<String> replacementInstances,
	Set<String> blockedInstances,
	Set<String> failedInstanceIps,
	String localDC)
	{
	return (failedInstanceIps.size() + blockedInstances.size()) <= (writeReplicas.size() - pendingReplicas.size() - 1);
	}

	@Override
	public boolean canBeSatisfied(Collection<? extends CassandraInstance> succeededInstances,
	ReplicationFactor replicationFactor,
	String localDC)
	{
	ensureNetworkTopologyStrategy(replicationFactor, LOCAL_ONE);
	Objects.requireNonNull(localDC, "localDC cannot be null");

	return succeededInstances.stream().anyMatch(instance -> instance.datacenter().equalsIgnoreCase(localDC));
	}
	};
	}
	}