Examples with SparkClientContext io.cdap.cdap.api.spark.SparkClientContext used on opensource projects

Example 1 with SparkClientContext

use of io.cdap.cdap.api.spark.SparkClientContext in project cdap by caskdata.

the class ETLSpark method initialize.

@Override
@TransactionPolicy(TransactionControl.EXPLICIT)
public void initialize() throws Exception {
    SparkClientContext context = getContext();
    SparkConf sparkConf = new SparkConf();
    sparkConf.set("spark.speculation", "false");
    // turn off auto-broadcast by default until we better understand the implications and can set this to a
    // value that we are confident is safe.
    sparkConf.set("spark.sql.autoBroadcastJoinThreshold", "-1");
    sparkConf.set("spark.maxRemoteBlockSizeFetchToMem", String.valueOf(Integer.MAX_VALUE - 512));
    sparkConf.set("spark.network.timeout", "600s");
    // Disable yarn app retries since spark already performs retries at a task level.
    sparkConf.set("spark.yarn.maxAppAttempts", "1");
    // to make sure fields that are the same but different casing are treated as different fields in auto-joins
    // see CDAP-17024
    sparkConf.set("spark.sql.caseSensitive", "true");
    context.setSparkConf(sparkConf);
    Map<String, String> properties = context.getSpecification().getProperties();
    BatchPhaseSpec phaseSpec = GSON.fromJson(properties.get(Constants.PIPELINEID), BatchPhaseSpec.class);
    for (Map.Entry<String, String> pipelineProperty : phaseSpec.getPipelineProperties().entrySet()) {
        sparkConf.set(pipelineProperty.getKey(), pipelineProperty.getValue());
    }
    PipelineRuntime pipelineRuntime = new PipelineRuntime(context);
    MacroEvaluator evaluator = new DefaultMacroEvaluator(pipelineRuntime.getArguments(), context.getLogicalStartTime(), context, context, context.getNamespace());
    SparkPreparer preparer = new SparkPreparer(context, context.getMetrics(), evaluator, pipelineRuntime);
    List<Finisher> finishers = preparer.prepare(phaseSpec);
    finisher = new CompositeFinisher(finishers);
}

Example 2 with SparkClientContext

use of io.cdap.cdap.api.spark.SparkClientContext in project cdap by caskdata.

the class DataStreamsSparkLauncher method initialize.

@TransactionPolicy(TransactionControl.EXPLICIT)
@Override
public void initialize() throws Exception {
    SparkClientContext context = getContext();
    String arguments = Joiner.on(", ").withKeyValueSeparator("=").join(context.getRuntimeArguments());
    WRAPPERLOGGER.info("Pipeline '{}' is started by user '{}' with arguments {}", context.getApplicationSpecification().getName(), UserGroupInformation.getCurrentUser().getShortUserName(), arguments);
    DataStreamsPipelineSpec spec = GSON.fromJson(context.getSpecification().getProperty(Constants.PIPELINEID), DataStreamsPipelineSpec.class);
    PipelinePluginContext pluginContext = new SparkPipelinePluginContext(context, context.getMetrics(), true, true);
    int numSources = 0;
    for (StageSpec stageSpec : spec.getStages()) {
        if (StreamingSource.PLUGIN_TYPE.equals(stageSpec.getPlugin().getType())) {
            StreamingSource<Object> streamingSource = pluginContext.newPluginInstance(stageSpec.getName());
            numSources = numSources + streamingSource.getRequiredExecutors();
        }
    }
    SparkConf sparkConf = new SparkConf();
    sparkConf.set("spark.streaming.backpressure.enabled", "true");
    sparkConf.set("spark.spark.streaming.blockInterval", String.valueOf(spec.getBatchIntervalMillis() / 5));
    sparkConf.set("spark.maxRemoteBlockSizeFetchToMem", String.valueOf(Integer.MAX_VALUE - 512));
    // spark... makes you set this to at least the number of receivers (streaming sources)
    // because it holds one thread per receiver, or one core in distributed mode.
    // so... we have to set this hacky master variable based on the isUnitTest setting in the config
    String extraOpts = spec.getExtraJavaOpts();
    if (extraOpts != null && !extraOpts.isEmpty()) {
        sparkConf.set("spark.driver.extraJavaOptions", extraOpts);
        sparkConf.set("spark.executor.extraJavaOptions", extraOpts);
    }
    // without this, stopping will hang on machines with few cores.
    sparkConf.set("spark.rpc.netty.dispatcher.numThreads", String.valueOf(numSources + 2));
    sparkConf.setMaster(String.format("local[%d]", numSources + 2));
    sparkConf.set("spark.executor.instances", String.valueOf(numSources + 2));
    if (spec.isUnitTest()) {
        sparkConf.setMaster(String.format("local[%d]", numSources + 1));
    }
    // override defaults with any user provided engine configs
    int minExecutors = numSources + 1;
    for (Map.Entry<String, String> property : spec.getProperties().entrySet()) {
        if ("spark.executor.instances".equals(property.getKey())) {
            // don't let the user set this to something that doesn't make sense
            try {
                int numExecutors = Integer.parseInt(property.getValue());
                if (numExecutors < minExecutors) {
                    LOG.warn("Number of executors {} is less than the minimum number required to run the pipeline. " + "Automatically increasing it to {}", numExecutors, minExecutors);
                    numExecutors = minExecutors;
                }
                sparkConf.set(property.getKey(), String.valueOf(numExecutors));
            } catch (NumberFormatException e) {
                throw new IllegalArgumentException("Number of spark executors was set to invalid value " + property.getValue(), e);
            }
        } else {
            sparkConf.set(property.getKey(), property.getValue());
        }
    }
    context.setSparkConf(sparkConf);
    WRAPPERLOGGER.info("Pipeline '{}' running", context.getApplicationSpecification().getName());
}

Example 3 with SparkClientContext

use of io.cdap.cdap.api.spark.SparkClientContext in project cdap by caskdata.

the class CharCountProgram method initialize.

@Override
public void initialize() throws Exception {
    SparkClientContext context = getContext();
    context.setSparkConf(new SparkConf().set("spark.io.compression.codec", "org.apache.spark.io.LZFCompressionCodec"));
    Table totals = context.getDataset("totals");
    totals.get(new Get("total").add("total")).getLong("total");
    totals.put(new Put("total").add("total", 0L));
}

Example 4 with SparkClientContext

use of io.cdap.cdap.api.spark.SparkClientContext in project cdap by caskdata.

the class ExternalSparkProgram method initialize.

@Override
protected void initialize() throws Exception {
    SparkClientContext context = getContext();
    BatchPhaseSpec phaseSpec = GSON.fromJson(getContext().getSpecification().getProperty(Constants.PIPELINEID), BatchPhaseSpec.class);
    SparkConf sparkConf = new SparkConf();
    for (Map.Entry<String, String> pipelineProperty : phaseSpec.getPipelineProperties().entrySet()) {
        sparkConf.set(pipelineProperty.getKey(), pipelineProperty.getValue());
    }
    context.setSparkConf(sparkConf);
    String stageName = context.getSpecification().getProperty(STAGE_NAME);
    Class<?> externalProgramClass = context.loadPluginClass(stageName);
    // If the external program implements Spark, instantiate it and call initialize() to provide full lifecycle support
    if (Spark.class.isAssignableFrom(externalProgramClass)) {
        MacroEvaluator macroEvaluator = new DefaultMacroEvaluator(new BasicArguments(context), context.getLogicalStartTime(), context, context, context.getNamespace());
        delegateSpark = context.newPluginInstance(stageName, macroEvaluator);
        if (delegateSpark instanceof AbstractSpark) {
            // noinspection unchecked
            ((AbstractSpark) delegateSpark).initialize(context);
        }
    }
}