-
Chen Chao authored
RT Author: Chen Chao <crazyjvm@gmail.com> Closes #114 from CrazyJvm/patch-1 and squashes the following commits: dcb0df5 [Chen Chao] maintain arbitrary state data for each key
Chen Chao authoredRT Author: Chen Chao <crazyjvm@gmail.com> Closes #114 from CrazyJvm/patch-1 and squashes the following commits: dcb0df5 [Chen Chao] maintain arbitrary state data for each key
layout: global
title: Spark Streaming Programming Guide- This will become a table of contents (this text will be scraped). {:toc}
Overview
Spark Streaming is an extension of the core Spark API that allows enables high-throughput,
fault-tolerant stream processing of live data streams. Data can be ingested from many sources
like Kafka, Flume, Twitter, ZeroMQ or plain old TCP sockets and be processed using complex
algorithms expressed with high-level functions like map, reduce, join and window.
Finally, processed data can be pushed out to filesystems, databases,
and live dashboards. In fact, you can apply Spark's in-built
machine learning algorithms, and
graph processing algorithms on data streams.
Internally, it works as follows. Spark Streaming receives live input data streams and divides the data into batches, which are then processed by the Spark engine to generate the final stream of results in batches.
Spark Streaming provides a high-level abstraction called discretized stream or DStream, which represents a continuous stream of data. DStreams can be created either from input data stream from sources such as Kafka and Flume, or by applying high-level operations on other DStreams. Internally, a DStream is represented as a sequence of RDDs.
This guide shows you how to start writing Spark Streaming programs with DStreams. You can write Spark Streaming programs in Scala or Java, both of which are presented in this guide. You will find tabs throughout this guide that let you choose between Scala and Java code snippets.
A Quick Example
Before we go into the details of how to write your own Spark Streaming program, let's take a quick look at what a simple Spark Streaming program looks like. Let's say we want to count the number of words in text data received from a data server listening on a TCP socket. All you need to do is as follows.
StreamingContext is the main entry point for all streaming functionality.
{% highlight scala %} import org.apache.spark.streaming._ import org.apache.spark.streaming.StreamingContext._ {% endhighlight %}
Then we create a StreamingContext object. Besides Spark's configuration, we specify that any DStream will be processed in 1 second batches.
{% highlight scala %} // Create a StreamingContext with a SparkConf configuration val ssc = new StreamingContext(sparkConf, Seconds(1)) {% endhighlight %}
Using this context, we then create a new DStream by specifying the IP address and port of the data server.
{% highlight scala %} // Create a DStream that will connect to serverIP:serverPort val lines = ssc.socketTextStream(serverIP, serverPort) {% endhighlight %}
This lines DStream represents the stream of data that will be received from the data
server. Each record in this DStream is a line of text. Next, we want to split the lines by
space into words.
{% highlight scala %} // Split each line into words val words = lines.flatMap(_.split(" ")) {% endhighlight %}
flatMap is a one-to-many DStream operation that creates a new DStream by
generating multiple new records from each record in the source DStream. In this case,
each line will be split into multiple words and the stream of words is represented as the
words DStream. Next, we want to count these words.
{% highlight scala %} // Count each word in each batch val pairs = words.map(word => (word, 1)) val wordCounts = pairs.reduceByKey(_ + _)
// Print a few of the counts to the console wordCounts.print() {% endhighlight %}
The words DStream is further mapped (one-to-one transformation) to a DStream of (word, 1) pairs, which is then reduced to get the frequency of words in each batch of data.
Finally, wordCounts.print() will print a few of the counts generated every second.
Note that when these lines are executed, Spark Streaming only sets up the computation it will perform when it is started, and no real processing has started yet. To start the processing after all the transformations have been setup, we finally call
{% highlight scala %} ssc.start() // Start the computation ssc.awaitTermination() // Wait for the computation to terminate {% endhighlight %}
The complete code can be found in the Spark Streaming example
NetworkWordCount.
First, we create a JavaStreamingContext object, which is the main entry point for all streaming functionality. Besides Spark's configuration, we specify that any DStream would be processed in 1 second batches.
{% highlight java %} // Create a StreamingContext with a SparkConf configuration JavaStreamingContext jssc = StreamingContext(sparkConf, new Duration(1000)) {% endhighlight %}
Using this context, we then create a new DStream by specifying the IP address and port of the data server.
{% highlight java %} // Create a DStream that will connect to serverIP:serverPort JavaDStream lines = jssc.socketTextStream(serverIP, serverPort); {% endhighlight %}
This lines DStream represents the stream of data that will be received from the data
server. Each record in this stream is a line of text. Then, we want to split the the lines by
space into words.
{% highlight java %} // Split each line into words JavaDStream words = lines.flatMap( new FlatMapFunction<String, String>() { @Override public Iterable call(String x) { return Lists.newArrayList(x.split(" ")); } }); {% endhighlight %}
flatMap is a DStream operation that creates a new DStream by
generating multiple new records from each record in the source DStream. In this case,
each line will be split into multiple words and the stream of words is represented as the
words DStream. Note that we defined the transformation using a
FlatMapFunction object.
As we will discover along the way, there are a number of such convenience classes in the Java API
that help define DStream transformations.
Next, we want to count these words.
{% highlight java %} // Count each word in each batch JavaPairDStream<String, Integer> pairs = words.map( new PairFunction<String, String, Integer>() { @Override public Tuple2<String, Integer> call(String s) throws Exception { return new Tuple2<String, Integer>(s, 1); } }); JavaPairDStream<String, Integer> wordCounts = pairs.reduceByKey( new Function2<Integer, Integer, Integer>() { @Override public Integer call(Integer i1, Integer i2) throws Exception { return i1 + i2; } }); wordCounts.print(); // Print a few of the counts to the console {% endhighlight %}
The words DStream is further mapped (one-to-one transformation) to a DStream of (word, 1) pairs, using a PairFunction
object. Then, it is reduced to get the frequency of words in each batch of data,
using a Function2 object.
Finally, wordCounts.print() will print a few of the counts generated every second.
Note that when these lines are executed, Spark Streaming only sets up the computation it will perform when it is started, and no real processing has started yet. To start the processing after all the transformations have been setup, we finally call
{% highlight java %} jssc.start(); // Start the computation jssc.awaitTermination(); // Wait for the computation to terminate {% endhighlight %}
The complete code can be found in the Spark Streaming example
JavaNetworkWordCount.
If you have already downloaded and built Spark, you can run this example as follows. You will first need to run Netcat (a small utility found in most Unix-like systems) as a data server by using
{% highlight bash %} $ nc -lk 9999 {% endhighlight %}
Then, in a different terminal, you can start the example by using
Then, any lines typed in the terminal running the netcat server will be counted and printed on screen every second. It will look something like this.
|
{% highlight bash %}
# TERMINAL 1:
# Running Netcat
$ nc -lk 9999 hello world ... {% endhighlight %} |
{% highlight bash %}
TERMINAL 2: RUNNING NetworkWordCount or JavaNetworkWordCount$ ./bin/run-example org.apache.spark.streaming.examples.NetworkWordCount local[2] localhost 9999 ...Time: 1357008430000 ms(hello,1) (world,1) ... {% endhighlight %} |
If you plan to run the Scala code for Spark Streaming-based use cases in the Spark shell, you should start the shell with the SparkConfiguration pre-configured to discard old batches periodically:
{% highlight bash %} $ SPARK_JAVA_OPTS=-Dspark.cleaner.ttl=10000 bin/spark-shell {% endhighlight %}
... and create your StreamingContext by wrapping the existing interactive shell
SparkContext object, sc:
{% highlight scala %} val ssc = new StreamingContext(sc, Seconds(1)) {% endhighlight %}
When working with the shell, you may also need to send a ^D to your netcat session
to force the pipeline to print the word counts to the console at the sink.
Basics
Next, we move beyond the simple example and elaborate on the basics of Spark Streaming that you need to know to write your streaming applications.
Linking
To write your own Spark Streaming program, you will have to add the following dependency to your SBT or Maven project:
groupId = org.apache.spark
artifactId = spark-streaming_{{site.SCALA_BINARY_VERSION}}
version = {{site.SPARK_VERSION}}For ingesting data from sources like Kafka and Flume that are not present in the Spark
Streaming core
API, you will have to add the corresponding
artifact spark-streaming-xyz_{{site.SCALA_BINARY_VERSION}} to the dependencies. For example,
some of the common ones are as follows.
| Source | Artifact |
|---|---|
| Kafka | spark-streaming-kafka_{{site.SCALA_BINARY_VERSION}} |
| Flume | spark-streaming-flume_{{site.SCALA_BINARY_VERSION}} |
| spark-streaming-twitter_{{site.SCALA_BINARY_VERSION}} | |
| ZeroMQ | spark-streaming-zeromq_{{site.SCALA_BINARY_VERSION}} |
| MQTT | spark-streaming-mqtt_{{site.SCALA_BINARY_VERSION}} |
For an up-to-date list, please refer to the Apache repository for the full list of supported sources and artifacts.
Initializing
To initialize a Spark Streaming program in Scala, a
StreamingContext
object has to be created, which is the main entry point of all Spark Streaming functionality.
A StreamingContext object can be created by using
{% highlight scala %} new StreamingContext(master, appName, batchDuration, [sparkHome], [jars]) {% endhighlight %}
To initialize a Spark Streaming program in Java, a
JavaStreamingContext
object has to be created, which is the main entry point of all Spark Streaming functionality.
A JavaStreamingContext object can be created by using
{% highlight scala %} new JavaStreamingContext(master, appName, batchInterval, [sparkHome], [jars]) {% endhighlight %}
The master parameter is a standard Spark cluster URL
and can be "local" for local testing. The appName is a name of your program,
which will be shown on your cluster's web UI. The batchInterval is the size of the batches,
as explained earlier. Finally, the last two parameters are needed to deploy your code to a cluster
if running in distributed mode, as described in the
Spark programming guide.
Additionally, the underlying SparkContext can be accessed as
streamingContext.sparkContext.
The batch interval must be set based on the latency requirements of your application and available cluster resources. See the Performance Tuning section for more details.
DStreams
Discretized Stream or DStream is the basic abstraction provided by Spark Streaming. It represents a continuous stream of data, either the input data stream received from source, or the processed data stream generated by transforming the input stream. Internally, it is represented by a continuous sequence of RDDs, which is Spark's abstraction of an immutable, distributed dataset. Each RDD in a DStream contains data from a certain interval, as shown in the following figure.
Any operation applied on a DStream translates to operations on the underlying RDDs. For example,
in the earlier example of converting a stream of lines to words,
the flatmap operation is applied on each RDD in the lines DStream to generate the RDDs of the
words DStream. This is shown the following figure.
These underlying RDD transformations are computed by the Spark engine. The DStream operations hide most of these details and provides the developer with higher-level API for convenience. These operations are discussed in detail in later sections.
Input Sources
We have already taken a look at the streamingContext.socketTextStream(...) in the quick
example which creates a DStream from text
data received over a TCP socket connection. Besides sockets, the core Spark Streaming API provides
methods for creating DStreams from files and Akka actors as input sources.
Specifically, for files, the DStream can be created as
Spark Streaming will monitor the directory dataDirectory for any Hadoop-compatible filesystem
and process any files created in that directory. Note that
- The files must have the same data format.
- The files must be created in the
dataDirectoryby atomically moving or renaming them into the data directory. - Once moved the files must not be changed.
For more details on streams from files, Akka actors and sockets, see the API documentations of the relevant functions in StreamingContext for Scala and JavaStreamingContext for Java.
Additional functionality for creating DStreams from sources such as Kafka, Flume, and Twitter
can be imported by adding the right dependencies as explained in an
earlier section. To take the
case of Kafka, after adding the artifact spark-streaming-kafka_{{site.SCALA_BINARY_VERSION}} to the
project dependencies, you can create a DStream from Kafka as
For more details on these additional sources, see the corresponding API documentation. Furthermore, you can also implement your own custom receiver for your sources. See the Custom Receiver Guide.
Operations
There are two kinds of DStream operations - transformations and output operations. Similar to RDD transformations, DStream transformations operate on one or more DStreams to create new DStreams with transformed data. After applying a sequence of transformations to the input streams, output operations need to called, which write data out to an external data sink, such as a filesystem or a database.
Transformations
DStreams support many of the transformations available on normal Spark RDD's. Some of the common ones are as follows.
| Transformation | Meaning |
|---|---|
| map(func) | Return a new DStream by passing each element of the source DStream through a function func. |
| flatMap(func) | Similar to map, but each input item can be mapped to 0 or more output items. |
| filter(func) | Return a new DStream by selecting only the records of the source DStream on which func returns true. |
| repartition(numPartitions) | Changes the level of parallelism in this DStream by creating more or fewer partitions. |
| union(otherStream) | Return a new DStream that contains the union of the elements in the source DStream and otherDStream. |
| count() | Return a new DStream of single-element RDDs by counting the number of elements in each RDD of the source DStream. |
| reduce(func) | Return a new DStream of single-element RDDs by aggregating the elements in each RDD of the source DStream using a function func (which takes two arguments and returns one). The function should be associative so that it can be computed in parallel. |
| countByValue() | When called on a DStream of elements of type K, return a new DStream of (K, Long) pairs where the value of each key is its frequency in each RDD of the source DStream. |
| reduceByKey(func, [numTasks]) | When called on a DStream of (K, V) pairs, return a new DStream of (K, V) pairs where the
values for each key are aggregated using the given reduce function. Note: By default,
this uses Spark's default number of parallel tasks (2 for local machine, 8 for a cluster) to
do the grouping. You can pass an optional numTasks argument to set a different
number of tasks. |
| join(otherStream, [numTasks]) | When called on two DStreams of (K, V) and (K, W) pairs, return a new DStream of (K, (V, W)) pairs with all pairs of elements for each key. |
| cogroup(otherStream, [numTasks]) | When called on DStream of (K, V) and (K, W) pairs, return a new DStream of (K, Seq[V], Seq[W]) tuples. |
| transform(func) | Return a new DStream by applying a RDD-to-RDD function to every RDD of the source DStream. This can be used to do arbitrary RDD operations on the DStream. |
| updateStateByKey(func) | Return a new "state" DStream where the state for each key is updated by applying the given function on the previous state of the key and the new values for the key. This can be used to maintain arbitrary state data for each key. |
The last two transformations are worth highlighting again.
UpdateStateByKey Operation
The updateStateByKey operation allows you to maintain arbitrary state while continuously updating
it with new information. To use this, you will have to do two steps.
- Define the state - The state can be of arbitrary data type.
- Define the state update function - Specify with a function how to update the state using the previous state and the new values from input stream.
Let's illustrate this with an example. Say you want to maintain a running count of each word seen in a text data stream. Here, the running count is the state and it is an integer. We define the update function as
{% highlight scala %} def updateFunction(newValues: Seq[Int], runningCount: Option[Int]): Option[Int] = { val newCount = ... // add the new values with the previous running count to get the new count Some(newCount) } {% endhighlight %}
This is applied on a DStream containing words (say, the pairs DStream containing (word, 1) pairs in the earlier example).
{% highlight scala %} val runningCounts = pairs.updateStateByKeyInt {% endhighlight %}
{% highlight java %} Function2<List, Optional, Optional> updateFunction = new Function2<List, Optional, Optional>() { @Override public Optional call(List values, Optional state) { Integer newSum = ... // add the new values with the previous running count to get the new count return Optional.of(newSum) } } {% endhighlight %}
This is applied on a DStream containing words (say, the pairs DStream containing (word, 1) pairs in the quick example).
{% highlight java %} JavaPairDStream<String, Integer> runningCounts = pairs.updateStateByKey(updateFunction); {% endhighlight %}
The update function will be called for each word, with newValues having a sequence of 1's (from
the (word, 1) pairs) and the runningCount having the previous count. For the complete
Scala code, take a look at the example
StatefulNetworkWordCount.
Transform Operation
The transform operation (along with its variations like transformWith) allows
arbitrary RDD-to-RDD functions to be applied on a DStream. It can be used to apply any RDD
operation that is not exposed in the DStream API.
For example, the functionality of joining every batch in a data stream
with another dataset is not directly exposed in the DStream API. However,
you can easily use transform to do this. This enables very powerful possibilities. For example,
if you want to do real-time data cleaning by joining the input data stream with precomputed
spam information (maybe generated with Spark as well) and then filtering based on it.
{% highlight scala %} val spamInfoRDD = sparkContext.hadoopFile(...) // RDD containing spam information
val cleanedDStream = inputDStream.transform(rdd => { rdd.join(spamInfoRDD).filter(...) // join data stream with spam information to do data cleaning ... }) {% endhighlight %}
{% highlight java %} // RDD containing spam information JavaPairRDD<String, Double> spamInfoRDD = javaSparkContext.hadoopFile(...);
JavaPairDStream<String, Integer> cleanedDStream = inputDStream.transform( new Function<JavaPairRDD<String, Integer>, JavaPairRDD<String, Integer>>() { @Override public JavaPairRDD<String, Integer> call(JavaPairRDD<String, Integer> rdd) throws Exception { rdd.join(spamInfoRDD).filter(...) // join data stream with spam information to do data cleaning ... } }); {% endhighlight %}
In fact, you can also use machine learning and
graph computation algorithms in the transform method.
Window Operations
Finally, Spark Streaming also provides windowed computations, which allow you to apply transformations over a sliding window of data. This following figure illustrates this sliding window.
As shown in the figure, every time the window slides over a source DStream, the source RDDs that fall within the window are combined and operated upon to produce the RDDs of the windowed DStream. In this specific case, the operation is applied over last 3 time units of data, and slides by 2 time units. This shows that any window-based operation needs to specify two parameters.
- window length - The duration of the window (3 in the figure)
- slide interval - The interval at which the window-based operation is performed (2 in the figure).
These two parameters must be multiples of the batch interval of the source DStream (1 in the figure).
Let's illustrate the window operations with an example. Say, you want to extend the
earlier example by generating word counts over last 30 seconds of data,
every 10 seconds. To do this, we have to apply the reduceByKey operation on the pairs DStream of
(word, 1) pairs over the last 30 seconds of data. This is done using the
operation reduceByKeyAndWindow.
{% highlight scala %} // Reduce last 30 seconds of data, every 10 seconds val windowedWordCounts = pairs.reduceByKeyAndWindow(_ + _, Seconds(30), Seconds(10)) {% endhighlight %}
{% highlight java %} // Reduce function adding two integers, defined separately for clarity Function2<Integer, Integer, Integer> reduceFunc = new Function2<Integer, Integer, Integer>() { @Override public Integer call(Integer i1, Integer i2) throws Exception { return i1 + i2; } };
// Reduce last 30 seconds of data, every 10 seconds JavaPairDStream<String, Integer> windowedWordCounts = pair.reduceByKeyAndWindow(reduceFunc, new Duration(30000), new Duration(10000)); {% endhighlight %}
Some of the common window-based operations are as follows. All of these operations take the said two parameters - windowLength and slideInterval.
| Transformation | Meaning |
|---|---|
| window(windowLength, slideInterval) | Return a new DStream which is computed based on windowed batches of the source DStream. |
| countByWindow(windowLength, slideInterval) | Return a sliding window count of elements in the stream. |
| reduceByWindow(func, windowLength, slideInterval) | Return a new single-element stream, created by aggregating elements in the stream over a sliding interval using func. The function should be associative so that it can be computed correctly in parallel. |
| reduceByKeyAndWindow(func, windowLength, slideInterval, [numTasks]) | When called on a DStream of (K, V) pairs, returns a new DStream of (K, V)
pairs where the values for each key are aggregated using the given reduce function func
over batches in a sliding window. Note: By default, this uses Spark's default number of
parallel tasks (2 for local machine, 8 for a cluster) to do the grouping. You can pass an optional
numTasks argument to set a different number of tasks.
|
| reduceByKeyAndWindow(func, invFunc, windowLength, slideInterval, [numTasks]) | A more efficient version of the above reduceByKeyAndWindow() where the reduce
value of each window is calculated incrementally using the reduce values of the previous window.
This is done by reducing the new data that enter the sliding window, and "inverse reducing" the
old data that leave the window. An example would be that of "adding" and "subtracting" counts
of keys as the window slides. However, it is applicable to only "invertible reduce functions",
that is, those reduce functions which have a corresponding "inverse reduce" function (taken as
parameter invFunc. Like in reduceByKeyAndWindow, the number of reduce tasks
is configurable through an optional argument.
|
| countByValueAndWindow(windowLength, slideInterval, [numTasks]) | When called on a DStream of (K, V) pairs, returns a new DStream of (K, Long) pairs where the
value of each key is its frequency within a sliding window. Like in
reduceByKeyAndWindow, the number of reduce tasks is configurable through an
optional argument.
|
Output Operations
When an output operator is called, it triggers the computation of a stream. Currently the following output operators are defined:
| Output Operation | Meaning |
|---|---|
| print() | Prints first ten elements of every batch of data in a DStream on the driver. |
| foreachRDD(func) | The fundamental output operator. Applies a function, func, to each RDD generated from the stream. This function should have side effects, such as printing output, saving the RDD to external files, or writing it over the network to an external system. |
| saveAsObjectFiles(prefix, [suffix]) | Save this DStream's contents as a SequenceFile of serialized objects. The file
name at each batch interval is generated based on prefix and
suffix: "prefix-TIME_IN_MS[.suffix]".
|
| saveAsTextFiles(prefix, [suffix]) | Save this DStream's contents as a text files. The file name at each batch interval is generated based on prefix and suffix: "prefix-TIME_IN_MS[.suffix]". |
| saveAsHadoopFiles(prefix, [suffix]) | Save this DStream's contents as a Hadoop file. The file name at each batch interval is generated based on prefix and suffix: "prefix-TIME_IN_MS[.suffix]". |
The complete list of DStream operations is available in the API documentation. For the Scala API, see DStream and PairDStreamFunctions. For the Java API, see JavaDStream and JavaPairDStream. Specifically for the Java API, see Spark's Java programming guide for more information.
Persistence
Similar to RDDs, DStreams also allow developers to persist the stream's data in memory. That is,
using persist() method on a DStream would automatically persist every RDD of that DStream in
memory. This is useful if the data in the DStream will be computed multiple times (e.g., multiple
operations on the same data). For window-based operations like reduceByWindow and
reduceByKeyAndWindow and state-based operations like updateStateByKey, this is implicitly true.
Hence, DStreams generated by window-based operations are automatically persisted in memory, without
the developer calling persist().
For input streams that receive data over the network (such as, Kafka, Flume, sockets, etc.), the default persistence level is set to replicate the data to two nodes for fault-tolerance.
Note that, unlike RDDs, the default persistence level of DStreams keeps the data serialized in memory. This is further discussed in the Performance Tuning section. More information on different persistence levels can be found in Spark Programming Guide.
RDD Checkpointing
A stateful operation is one which operates over multiple batches of data. This includes all
window-based operations and the updateStateByKey operation. Since stateful operations have a
dependency on previous batches of data, they continuously accumulate metadata over time.
To clear this metadata, streaming supports periodic checkpointing by saving intermediate data
to HDFS. Note that checkpointing also incurs the cost of saving to HDFS which may cause the
corresponding batch to take longer to process. Hence, the interval of checkpointing needs to be
set carefully. At small batch sizes (say 1 second), checkpointing every batch may significantly
reduce operation throughput. Conversely, checkpointing too slowly causes the lineage and task
sizes to grow which may have detrimental effects. Typically, a checkpoint interval of 5 - 10
times of sliding interval of a DStream is good setting to try.
To enable checkpointing, the developer has to provide the HDFS path to which RDD will be saved. This is done by using
{% highlight scala %} ssc.checkpoint(hdfsPath) // assuming ssc is the StreamingContext or JavaStreamingContext {% endhighlight %}
The interval of checkpointing of a DStream can be set by using
{% highlight scala %} dstream.checkpoint(checkpointInterval) {% endhighlight %}
For DStreams that must be checkpointed (that is, DStreams created by updateStateByKey and
reduceByKeyAndWindow with inverse function), the checkpoint interval of the DStream is by
default set to a multiple of the DStream's sliding interval such that its at least 10 seconds.
Performance Tuning
Getting the best performance of a Spark Streaming application on a cluster requires a bit of tuning. This section explains a number of the parameters and configurations that can tuned to improve the performance of you application. At a high level, you need to consider two things:
- Reducing the processing time of each batch of data by efficiently using cluster resources.
- Setting the right batch size such that the data processing can keep up with the data ingestion.
Reducing the Processing Time of each Batch
There are a number of optimizations that can be done in Spark to minimize the processing time of each batch. These have been discussed in detail in Tuning Guide. This section highlights some of the most important ones.
Level of Parallelism
Cluster resources maybe under-utilized if the number of parallel tasks used in any stage of the
computation is not high enough. For example, for distributed reduce operations like reduceByKey
and reduceByKeyAndWindow, the default number of parallel tasks is 8. You can pass the level of
parallelism as an argument (see the
PairDStreamFunctions
documentation), or set the config property
spark.default.parallelism to change the default.
Data Serialization
The overhead of data serialization can be significant, especially when sub-second batch sizes are to be achieved. There are two aspects to it.
-
Serialization of RDD data in Spark: Please refer to the detailed discussion on data serialization in the Tuning Guide. However, note that unlike Spark, by default RDDs are persisted as serialized byte arrays to minimize pauses related to GC.
-
Serialization of input data: To ingest external data into Spark, data received as bytes (say, from the network) needs to deserialized from bytes and re-serialized into Spark's serialization format. Hence, the deserialization overhead of input data may be a bottleneck.
Task Launching Overheads
If the number of tasks launched per second is high (say, 50 or more per second), then the overhead of sending out tasks to the slaves maybe significant and will make it hard to achieve sub-second latencies. The overhead can be reduced by the following changes:
-
Task Serialization: Using Kryo serialization for serializing tasks can reduced the task sizes, and therefore reduce the time taken to send them to the slaves.
-
Execution mode: Running Spark in Standalone mode or coarse-grained Mesos mode leads to better task launch times than the fine-grained Mesos mode. Please refer to the Running on Mesos guide for more details.
These changes may reduce batch processing time by 100s of milliseconds, thus allowing sub-second batch size to be viable.
Setting the Right Batch Size
For a Spark Streaming application running on a cluster to be stable, the processing of the data streams must keep up with the rate of ingestion of the data streams. Depending on the type of computation, the batch size used may have significant impact on the rate of ingestion that can be sustained by the Spark Streaming application on a fixed cluster resources. For example, let us consider the earlier WordCountNetwork example. For a particular data rate, the system may be able to keep up with reporting word counts every 2 seconds (i.e., batch size of 2 seconds), but not every 500 milliseconds.
A good approach to figure out the right batch size for your application is to test it with a conservative batch size (say, 5-10 seconds) and a low data rate. To verify whether the system is able to keep up with data rate, you can check the value of the end-to-end delay experienced by each processed batch (either look for "Total delay" in Spark driver log4j logs, or use the StreamingListener interface). If the delay is maintained to be comparable to the batch size, then system is stable. Otherwise, if the delay is continuously increasing, it means that the system is unable to keep up and it therefore unstable. Once you have an idea of a stable configuration, you can try increasing the data rate and/or reducing the batch size. Note that momentary increase in the delay due to temporary data rate increases maybe fine as long as the delay reduces back to a low value (i.e., less than batch size).
24/7 Operation
By default, Spark does not forget any of the metadata (RDDs generated, stages processed, etc.).
But for a Spark Streaming application to operate 24/7, it is necessary for Spark to do periodic
cleanup of it metadata. This can be enabled by setting the
configuration property spark.cleaner.ttl to the number of
seconds you want any metadata to persist. For example, setting spark.cleaner.ttl to 600 would
cause Spark periodically cleanup all metadata and persisted RDDs that are older than 10 minutes.
Note, that this property needs to be set before the SparkContext is created.
This value is closely tied with any window operation that is being used. Any window operation would require the input data to be persisted in memory for at least the duration of the window. Hence it is necessary to set the delay to at least the value of the largest window operation used in the Spark Streaming application. If this delay is set too low, the application will throw an exception saying so.
Monitoring
Besides Spark's in-built monitoring capabilities, the progress of a Spark Streaming program can also be monitored using the [StreamingListener] (api/streaming/index.html#org.apache.spark.scheduler.StreamingListener) interface, which allows you to get statistics of batch processing times, queueing delays, and total end-to-end delays. Note that this is still an experimental API and it is likely to be improved upon (i.e., more information reported) in the future.
Memory Tuning
Tuning the memory usage and GC behavior of Spark applications have been discussed in great detail in the Tuning Guide. It is recommended that you read that. In this section, we highlight a few customizations that are strongly recommended to minimize GC related pauses in Spark Streaming applications and achieving more consistent batch processing times.
-
Default persistence level of DStreams: Unlike RDDs, the default persistence level of DStreams serializes the data in memory (that is, StorageLevel.MEMORY_ONLY_SER for DStream compared to StorageLevel.MEMORY_ONLY for RDDs). Even though keeping the data serialized incurs higher serialization/deserialization overheads, it significantly reduces GC pauses.
-
Clearing persistent RDDs: By default, all persistent RDDs generated by Spark Streaming will be cleared from memory based on Spark's in-built policy (LRU). If
spark.cleaner.ttlis set, then persistent RDDs that are older than that value are periodically cleared. As mentioned earlier, this needs to be careful set based on operations used in the Spark Streaming program. However, a smarter unpersisting of RDDs can be enabled by setting the configuration propertyspark.streaming.unpersisttotrue. This makes the system to figure out which RDDs are not necessary to be kept around and unpersists them. This is likely to reduce the RDD memory usage of Spark, potentially improving GC behavior as well. -
Concurrent garbage collector: Using the concurrent mark-and-sweep GC further minimizes the variability of GC pauses. Even though concurrent GC is known to reduce the overall processing throughput of the system, its use is still recommended to achieve more consistent batch processing times.
Fault-tolerance Properties
In this section, we are going to discuss the behavior of Spark Streaming application in the event of a node failure. To understand this, let us remember the basic fault-tolerance properties of Spark's RDDs.
- An RDD is an immutable, deterministically re-computable, distributed dataset. Each RDD remembers the lineage of deterministic operations that were used on a fault-tolerant input dataset to create it.
- If any partition of an RDD is lost due to a worker node failure, then that partition can be re-computed from the original fault-tolerant dataset using the lineage of operations.
Since all data transformations in Spark Streaming are based on RDD operations, as long as the input dataset is present, all intermediate data can recomputed. Keeping these properties in mind, we are going to discuss the failure semantics in more detail.
Failure of a Worker Node
There are two failure behaviors based on which input sources are used.
- Using HDFS files as input source - Since the data is reliably stored on HDFS, all data can re-computed and therefore no data will be lost due to any failure.
- Using any input source that receives data through a network - For network-based data sources like Kafka and Flume, the received input data is replicated in memory between nodes of the cluster (default replication factor is 2). So if a worker node fails, then the system can recompute the lost from the the left over copy of the input data. However, if the worker node where a network receiver was running fails, then a tiny bit of data may be lost, that is, the data received by the system but not yet replicated to other node(s). The receiver will be started on a different node and it will continue to receive data.
Since all data is modeled as RDDs with their lineage of deterministic operations, any recomputation
always leads to the same result. As a result, all DStream transformations are guaranteed to have
exactly-once semantics. That is, the final transformed result will be same even if there were
was a worker node failure. However, output operations (like foreachRDD) have at-least once
semantics, that is, the transformed data may get written to an external entity more than once in
the event of a worker failure. While this is acceptable for saving to HDFS using the
saveAs*Files operations (as the file will simply get over-written by the same data),
additional transactions-like mechanisms may be necessary to achieve exactly-once semantics
for output operations.
Failure of the Driver Node
For a streaming application to operate 24/7, Spark Streaming allows a streaming computation
to be resumed even after the failure of the driver node. Spark Streaming periodically writes the
metadata information of the DStreams setup through the StreamingContext to a
HDFS directory (can be any Hadoop-compatible filesystem). This periodic
checkpointing can be enabled by setting the checkpoint
directory using ssc.checkpoint(<checkpoint directory>) as described
earlier. On failure of the driver node,
the lost StreamingContext can be recovered from this information, and restarted.
To allow a Spark Streaming program to be recoverable, it must be written in a way such that it has the following behavior:
- When the program is being started for the first time, it will create a new StreamingContext, set up all the streams and then call start().
- When the program is being restarted after failure, it will re-create a StreamingContext from the checkpoint data in the checkpoint directory.
This behavior is made simple by using StreamingContext.getOrCreate. This is used as follows.
{% highlight scala %} // Function to create and setup a new StreamingContext def functionToCreateContext(): StreamingContext = { val ssc = new StreamingContext(...) // new context val lines = ssc.socketTextStream(...) // create DStreams ... ssc.checkpoint(checkpointDirectory) // set checkpoint directory ssc }
// Get StreaminContext from checkpoint data or create a new one val context = StreamingContext.getOrCreate(checkpointDirectory, functionToCreateContext _)
// Do additional setup on context that needs to be done, // irrespective of whether it is being started or restarted context. ...
// Start the context context.start() context.awaitTermination() {% endhighlight %}
If the checkpointDirectory exists, then the context will be recreated from the checkpoint data.
If the directory does not exist (i.e., running for the first time),
then the function functionToCreateContext will be called to create a new
context and set up the DStreams. See the Scala example
RecoverableNetworkWordCount.
This example appends the word counts of network data into a file.
You can also explicitly create a StreamingContext from the checkpoint data and start the
computation by using new StreamingContext(checkpointDirectory).
This behavior is made simple by using JavaStreamingContext.getOrCreate. This is used as follows.
{% highlight java %} // Create a factory object that can create a and setup a new JavaStreamingContext JavaStreamingContextFactory contextFactory = new JavaStreamingContextFactory() { JavaStreamingContextFactory create() { JavaStreamingContext jssc = new JavaStreamingContext(...); // new context JavaDStream lines = jssc.socketTextStream(...); // create DStreams ... jssc.checkpoint(checkpointDirectory); // set checkpoint directory return jssc; } };
// Get JavaStreamingContext from checkpoint data or create a new one JavaStreamingContext context = JavaStreamingContext.getOrCreate(checkpointDirectory, contextFactory);
// Do additional setup on context that needs to be done, // irrespective of whether it is being started or restarted context. ...
// Start the context context.start(); context.awaitTermination(); {% endhighlight %}
If the checkpointDirectory exists, then the context will be recreated from the checkpoint data.
If the directory does not exist (i.e., running for the first time),
then the function contextFactory will be called to create a new
context and set up the DStreams. See the Scala example
JavaRecoverableWordCount
(note that this example is missing in the 0.9 release, so you can test it using the master branch).
This example appends the word counts of network data into a file.
You can also explicitly create a JavaStreamingContext from the checkpoint data and start
the computation by using new JavaStreamingContext(checkpointDirectory).
Note: If Spark Streaming and/or the Spark Streaming program is recompiled,
you must create a new StreamingContext or JavaStreamingContext,
not recreate from checkpoint data. This is because trying to load a
context from checkpoint data may fail if the data was generated before recompilation of the
classes. So, if you are using getOrCreate, then make sure that the checkpoint directory is
explicitly deleted every time recompiled code needs to be launched.
This failure recovery can be done automatically using Spark's
standalone cluster mode, which allows the driver of any Spark application
to be launched within the cluster and be restarted on failure (see
supervise mode). This can be
tested locally by launching the above example using the supervise mode in a
local standalone cluster and killing the java process running the driver (will be shown as
DriverWrapper when jps is run to show all active Java processes). The driver should be
automatically restarted, and the word counts will cont
For other deployment environments like Mesos and Yarn, you have to restart the driver through other mechanisms.
Recovery Semantics
There are two different failure behaviors based on which input sources are used.
- Using HDFS files as input source - Since the data is reliably stored on HDFS, all data can re-computed and therefore no data will be lost due to any failure.
-
Using any input source that receives data through a network - The received input data is
replicated in memory to multiple nodes. Since all the data in the Spark worker's memory is lost
when the Spark driver fails, the past input data will not be accessible and driver recovers.
Hence, if stateful and window-based operations are used
(like
updateStateByKey,window,countByValueAndWindow, etc.), then the intermediate state will not be recovered completely.
In future releases, we will support full recoverability for all input sources. Note that for
non-stateful transformations like map, count, and reduceByKey, with all input streams,
the system, upon restarting, will continue to receive and process new data.
To better understand the behavior of the system under driver failure with a HDFS source, let's consider what will happen with a file input stream. Specifically, in the case of the file input stream, it will correctly identify new files that were created while the driver was down and process them in the same way as it would have if the driver had not failed. To explain further in the case of file input stream, we shall use an example. Let's say, files are being generated every second, and a Spark Streaming program reads every new file and output the number of lines in the file. This is what the sequence of outputs would be with and without a driver failure.
| Time | Number of lines in input file | Output without driver failure | Output with driver failure |
|---|---|---|---|
| 1 | 10 | 10 | 10 |
| 2 | 20 | 20 | 20 |
| 3 | 30 | 30 | 30 |
| 4 | 40 | 40 | [DRIVER FAILS] no output |
| 5 | 50 | 50 | no output |
| 6 | 60 | 60 | no output |
| 7 | 70 | 70 | [DRIVER RECOVERS] 40, 50, 60, 70 |
| 8 | 80 | 80 | 80 |
| 9 | 90 | 90 | 90 |
| 10 | 100 | 100 | 100 |
If the driver had crashed in the middle of the processing of time 3, then it will process time 3 and output 30 after recovery.