Search in sources :

Example 1 with TaskAttemptDescriptor

use of org.apache.hyracks.control.common.job.TaskAttemptDescriptor in project asterixdb by apache.

the class StartTasksWork method run.

@Override
public void run() {
    Task task = null;
    try {
        NCServiceContext serviceCtx = ncs.getContext();
        Joblet joblet = getOrCreateLocalJoblet(deploymentId, jobId, serviceCtx, acgBytes);
        final ActivityClusterGraph acg = joblet.getActivityClusterGraph();
        IRecordDescriptorProvider rdp = new IRecordDescriptorProvider() {

            @Override
            public RecordDescriptor getOutputRecordDescriptor(ActivityId aid, int outputIndex) {
                ActivityCluster ac = acg.getActivityMap().get(aid);
                IConnectorDescriptor conn = ac.getActivityOutputMap().get(aid).get(outputIndex);
                return ac.getConnectorRecordDescriptorMap().get(conn.getConnectorId());
            }

            @Override
            public RecordDescriptor getInputRecordDescriptor(ActivityId aid, int inputIndex) {
                ActivityCluster ac = acg.getActivityMap().get(aid);
                IConnectorDescriptor conn = ac.getActivityInputMap().get(aid).get(inputIndex);
                return ac.getConnectorRecordDescriptorMap().get(conn.getConnectorId());
            }
        };
        for (TaskAttemptDescriptor td : taskDescriptors) {
            TaskAttemptId taId = td.getTaskAttemptId();
            TaskId tid = taId.getTaskId();
            ActivityId aid = tid.getActivityId();
            ActivityCluster ac = acg.getActivityMap().get(aid);
            IActivity han = ac.getActivityMap().get(aid);
            if (LOGGER.isLoggable(Level.INFO)) {
                LOGGER.info("Initializing " + taId + " -> " + han);
            }
            final int partition = tid.getPartition();
            List<IConnectorDescriptor> inputs = ac.getActivityInputMap().get(aid);
            task = new Task(joblet, taId, han.getClass().getName(), ncs.getExecutor(), ncs, createInputChannels(td, inputs));
            IOperatorNodePushable operator = han.createPushRuntime(task, rdp, partition, td.getPartitionCount());
            List<IPartitionCollector> collectors = new ArrayList<>();
            if (inputs != null) {
                for (int i = 0; i < inputs.size(); ++i) {
                    IConnectorDescriptor conn = inputs.get(i);
                    IConnectorPolicy cPolicy = connectorPoliciesMap.get(conn.getConnectorId());
                    if (LOGGER.isLoggable(Level.INFO)) {
                        LOGGER.info("input: " + i + ": " + conn.getConnectorId());
                    }
                    RecordDescriptor recordDesc = ac.getConnectorRecordDescriptorMap().get(conn.getConnectorId());
                    IPartitionCollector collector = createPartitionCollector(td, partition, task, i, conn, recordDesc, cPolicy);
                    collectors.add(collector);
                }
            }
            List<IConnectorDescriptor> outputs = ac.getActivityOutputMap().get(aid);
            if (outputs != null) {
                for (int i = 0; i < outputs.size(); ++i) {
                    final IConnectorDescriptor conn = outputs.get(i);
                    RecordDescriptor recordDesc = ac.getConnectorRecordDescriptorMap().get(conn.getConnectorId());
                    IConnectorPolicy cPolicy = connectorPoliciesMap.get(conn.getConnectorId());
                    IPartitionWriterFactory pwFactory = createPartitionWriterFactory(task, cPolicy, jobId, conn, partition, taId, flags);
                    if (LOGGER.isLoggable(Level.INFO)) {
                        LOGGER.info("output: " + i + ": " + conn.getConnectorId());
                    }
                    IFrameWriter writer = conn.createPartitioner(task, recordDesc, pwFactory, partition, td.getPartitionCount(), td.getOutputPartitionCounts()[i]);
                    operator.setOutputFrameWriter(i, writer, recordDesc);
                }
            }
            task.setTaskRuntime(collectors.toArray(new IPartitionCollector[collectors.size()]), operator);
            joblet.addTask(task);
            task.start();
        }
    } catch (Exception e) {
        LOGGER.log(Level.WARNING, "Failure starting a task", e);
        // notify cc of start task failure
        List<Exception> exceptions = new ArrayList<>();
        ExceptionUtils.setNodeIds(exceptions, ncs.getId());
        ncs.getWorkQueue().schedule(new NotifyTaskFailureWork(ncs, task, exceptions));
    }
}
Also used : IFrameWriter(org.apache.hyracks.api.comm.IFrameWriter) Task(org.apache.hyracks.control.nc.Task) TaskId(org.apache.hyracks.api.dataflow.TaskId) RecordDescriptor(org.apache.hyracks.api.dataflow.value.RecordDescriptor) ActivityId(org.apache.hyracks.api.dataflow.ActivityId) ArrayList(java.util.ArrayList) IRecordDescriptorProvider(org.apache.hyracks.api.dataflow.value.IRecordDescriptorProvider) Joblet(org.apache.hyracks.control.nc.Joblet) IActivity(org.apache.hyracks.api.dataflow.IActivity) NCServiceContext(org.apache.hyracks.control.nc.application.NCServiceContext) INCServiceContext(org.apache.hyracks.api.application.INCServiceContext) List(java.util.List) ArrayList(java.util.ArrayList) IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) IPartitionCollector(org.apache.hyracks.api.comm.IPartitionCollector) TaskAttemptId(org.apache.hyracks.api.dataflow.TaskAttemptId) IConnectorPolicy(org.apache.hyracks.api.dataflow.connectors.IConnectorPolicy) IPartitionWriterFactory(org.apache.hyracks.api.comm.IPartitionWriterFactory) HyracksDataException(org.apache.hyracks.api.exceptions.HyracksDataException) UnknownHostException(java.net.UnknownHostException) HyracksException(org.apache.hyracks.api.exceptions.HyracksException) ActivityCluster(org.apache.hyracks.api.job.ActivityCluster) TaskAttemptDescriptor(org.apache.hyracks.control.common.job.TaskAttemptDescriptor) ActivityClusterGraph(org.apache.hyracks.api.job.ActivityClusterGraph) IOperatorNodePushable(org.apache.hyracks.api.dataflow.IOperatorNodePushable)

Example 2 with TaskAttemptDescriptor

use of org.apache.hyracks.control.common.job.TaskAttemptDescriptor in project asterixdb by apache.

the class JobExecutor method startTasks.

private void startTasks(Map<String, List<TaskAttemptDescriptor>> taskAttemptMap) throws HyracksException {
    final DeploymentId deploymentId = jobRun.getDeploymentId();
    final JobId jobId = jobRun.getJobId();
    final ActivityClusterGraph acg = jobRun.getActivityClusterGraph();
    final Map<ConnectorDescriptorId, IConnectorPolicy> connectorPolicies = new HashMap<>(jobRun.getConnectorPolicyMap());
    INodeManager nodeManager = ccs.getNodeManager();
    try {
        byte[] acgBytes = predistributed ? null : JavaSerializationUtils.serialize(acg);
        for (Map.Entry<String, List<TaskAttemptDescriptor>> entry : taskAttemptMap.entrySet()) {
            String nodeId = entry.getKey();
            final List<TaskAttemptDescriptor> taskDescriptors = entry.getValue();
            final NodeControllerState node = nodeManager.getNodeControllerState(nodeId);
            if (node != null) {
                node.getActiveJobIds().add(jobRun.getJobId());
                boolean changed = jobRun.getParticipatingNodeIds().add(nodeId);
                if (LOGGER.isLoggable(Level.FINE)) {
                    LOGGER.fine("Starting: " + taskDescriptors + " at " + entry.getKey());
                }
                byte[] jagBytes = changed ? acgBytes : null;
                node.getNodeController().startTasks(deploymentId, jobId, jagBytes, taskDescriptors, connectorPolicies, jobRun.getFlags());
            }
        }
    } catch (Exception e) {
        throw new HyracksException(e);
    }
}
Also used : INodeManager(org.apache.hyracks.control.cc.cluster.INodeManager) DeploymentId(org.apache.hyracks.api.deployment.DeploymentId) HashMap(java.util.HashMap) IConnectorPolicy(org.apache.hyracks.api.dataflow.connectors.IConnectorPolicy) ConnectorDescriptorId(org.apache.hyracks.api.dataflow.ConnectorDescriptorId) HyracksException(org.apache.hyracks.api.exceptions.HyracksException) HyracksException(org.apache.hyracks.api.exceptions.HyracksException) TaskAttemptDescriptor(org.apache.hyracks.control.common.job.TaskAttemptDescriptor) ActivityClusterGraph(org.apache.hyracks.api.job.ActivityClusterGraph) ArrayList(java.util.ArrayList) List(java.util.List) NodeControllerState(org.apache.hyracks.control.cc.NodeControllerState) HashMap(java.util.HashMap) Map(java.util.Map) JobId(org.apache.hyracks.api.job.JobId)

Example 3 with TaskAttemptDescriptor

use of org.apache.hyracks.control.common.job.TaskAttemptDescriptor in project asterixdb by apache.

the class JobExecutor method assignTaskLocations.

private void assignTaskLocations(TaskCluster tc, Map<String, List<TaskAttemptDescriptor>> taskAttemptMap) throws HyracksException {
    ActivityClusterGraph acg = jobRun.getActivityClusterGraph();
    Task[] tasks = tc.getTasks();
    List<TaskClusterAttempt> tcAttempts = tc.getAttempts();
    int attempts = tcAttempts.size();
    TaskClusterAttempt tcAttempt = new TaskClusterAttempt(tc, attempts);
    Map<TaskId, TaskAttempt> taskAttempts = new HashMap<>();
    Map<TaskId, LValueConstraintExpression> locationMap = new HashMap<>();
    for (int i = 0; i < tasks.length; ++i) {
        Task ts = tasks[i];
        TaskId tid = ts.getTaskId();
        TaskAttempt taskAttempt = new TaskAttempt(tcAttempt, new TaskAttemptId(new TaskId(tid.getActivityId(), tid.getPartition()), attempts), ts);
        taskAttempt.setStatus(TaskAttempt.TaskStatus.INITIALIZED, null);
        locationMap.put(tid, new PartitionLocationExpression(tid.getActivityId().getOperatorDescriptorId(), tid.getPartition()));
        taskAttempts.put(tid, taskAttempt);
    }
    tcAttempt.setTaskAttempts(taskAttempts);
    solver.solve(locationMap.values());
    for (int i = 0; i < tasks.length; ++i) {
        Task ts = tasks[i];
        TaskId tid = ts.getTaskId();
        TaskAttempt taskAttempt = taskAttempts.get(tid);
        String nodeId = assignLocation(acg, locationMap, tid, taskAttempt);
        taskAttempt.setNodeId(nodeId);
        taskAttempt.setStatus(TaskAttempt.TaskStatus.RUNNING, null);
        taskAttempt.setStartTime(System.currentTimeMillis());
        List<TaskAttemptDescriptor> tads = taskAttemptMap.get(nodeId);
        if (tads == null) {
            tads = new ArrayList<>();
            taskAttemptMap.put(nodeId, tads);
        }
        OperatorDescriptorId opId = tid.getActivityId().getOperatorDescriptorId();
        jobRun.registerOperatorLocation(opId, tid.getPartition(), nodeId);
        ActivityPartitionDetails apd = ts.getActivityPlan().getActivityPartitionDetails();
        TaskAttemptDescriptor tad = new TaskAttemptDescriptor(taskAttempt.getTaskAttemptId(), apd.getPartitionCount(), apd.getInputPartitionCounts(), apd.getOutputPartitionCounts());
        tads.add(tad);
    }
    tcAttempt.initializePendingTaskCounter();
    tcAttempts.add(tcAttempt);
    /**
         * Improvement for reducing master/slave message communications, for each TaskAttemptDescriptor,
         * we set the NetworkAddress[][] partitionLocations, in which each row is for an incoming connector descriptor
         * and each column is for an input channel of the connector.
         */
    INodeManager nodeManager = ccs.getNodeManager();
    for (Map.Entry<String, List<TaskAttemptDescriptor>> e : taskAttemptMap.entrySet()) {
        List<TaskAttemptDescriptor> tads = e.getValue();
        for (TaskAttemptDescriptor tad : tads) {
            TaskAttemptId taid = tad.getTaskAttemptId();
            int attempt = taid.getAttempt();
            TaskId tid = taid.getTaskId();
            ActivityId aid = tid.getActivityId();
            List<IConnectorDescriptor> inConnectors = acg.getActivityInputs(aid);
            int[] inPartitionCounts = tad.getInputPartitionCounts();
            if (inPartitionCounts == null) {
                continue;
            }
            NetworkAddress[][] partitionLocations = new NetworkAddress[inPartitionCounts.length][];
            for (int i = 0; i < inPartitionCounts.length; ++i) {
                ConnectorDescriptorId cdId = inConnectors.get(i).getConnectorId();
                IConnectorPolicy policy = jobRun.getConnectorPolicyMap().get(cdId);
                /**
                     * carry sender location information into a task
                     * when it is not the case that it is an re-attempt and the send-side
                     * is materialized blocking.
                     */
                if (attempt > 0 && policy.materializeOnSendSide() && policy.consumerWaitsForProducerToFinish()) {
                    continue;
                }
                ActivityId producerAid = acg.getProducerActivity(cdId);
                partitionLocations[i] = new NetworkAddress[inPartitionCounts[i]];
                for (int j = 0; j < inPartitionCounts[i]; ++j) {
                    TaskId producerTaskId = new TaskId(producerAid, j);
                    String nodeId = findTaskLocation(producerTaskId);
                    partitionLocations[i][j] = nodeManager.getNodeControllerState(nodeId).getDataPort();
                }
            }
            tad.setInputPartitionLocations(partitionLocations);
        }
    }
    tcAttempt.setStatus(TaskClusterAttempt.TaskClusterStatus.RUNNING);
    tcAttempt.setStartTime(System.currentTimeMillis());
    inProgressTaskClusters.add(tc);
}
Also used : INodeManager(org.apache.hyracks.control.cc.cluster.INodeManager) Task(org.apache.hyracks.control.cc.job.Task) TaskId(org.apache.hyracks.api.dataflow.TaskId) TaskClusterAttempt(org.apache.hyracks.control.cc.job.TaskClusterAttempt) HashMap(java.util.HashMap) ActivityId(org.apache.hyracks.api.dataflow.ActivityId) ConnectorDescriptorId(org.apache.hyracks.api.dataflow.ConnectorDescriptorId) NetworkAddress(org.apache.hyracks.api.comm.NetworkAddress) ArrayList(java.util.ArrayList) List(java.util.List) TaskAttempt(org.apache.hyracks.control.cc.job.TaskAttempt) PartitionLocationExpression(org.apache.hyracks.api.constraints.expressions.PartitionLocationExpression) IConnectorDescriptor(org.apache.hyracks.api.dataflow.IConnectorDescriptor) OperatorDescriptorId(org.apache.hyracks.api.dataflow.OperatorDescriptorId) TaskAttemptId(org.apache.hyracks.api.dataflow.TaskAttemptId) IConnectorPolicy(org.apache.hyracks.api.dataflow.connectors.IConnectorPolicy) Constraint(org.apache.hyracks.api.constraints.Constraint) LValueConstraintExpression(org.apache.hyracks.api.constraints.expressions.LValueConstraintExpression) TaskAttemptDescriptor(org.apache.hyracks.control.common.job.TaskAttemptDescriptor) ActivityClusterGraph(org.apache.hyracks.api.job.ActivityClusterGraph) HashMap(java.util.HashMap) Map(java.util.Map)

Aggregations

ArrayList (java.util.ArrayList)3 List (java.util.List)3 IConnectorPolicy (org.apache.hyracks.api.dataflow.connectors.IConnectorPolicy)3 ActivityClusterGraph (org.apache.hyracks.api.job.ActivityClusterGraph)3 TaskAttemptDescriptor (org.apache.hyracks.control.common.job.TaskAttemptDescriptor)3 HashMap (java.util.HashMap)2 Map (java.util.Map)2 ActivityId (org.apache.hyracks.api.dataflow.ActivityId)2 ConnectorDescriptorId (org.apache.hyracks.api.dataflow.ConnectorDescriptorId)2 IConnectorDescriptor (org.apache.hyracks.api.dataflow.IConnectorDescriptor)2 TaskAttemptId (org.apache.hyracks.api.dataflow.TaskAttemptId)2 TaskId (org.apache.hyracks.api.dataflow.TaskId)2 HyracksException (org.apache.hyracks.api.exceptions.HyracksException)2 INodeManager (org.apache.hyracks.control.cc.cluster.INodeManager)2 UnknownHostException (java.net.UnknownHostException)1 INCServiceContext (org.apache.hyracks.api.application.INCServiceContext)1 IFrameWriter (org.apache.hyracks.api.comm.IFrameWriter)1 IPartitionCollector (org.apache.hyracks.api.comm.IPartitionCollector)1 IPartitionWriterFactory (org.apache.hyracks.api.comm.IPartitionWriterFactory)1 NetworkAddress (org.apache.hyracks.api.comm.NetworkAddress)1