Discussion thread | https://lists.apache.org/thread/qq39rmg3f23ysx5m094s4c4cq0m4tdj5 |
---|---|
Vote thread | - |
JIRA |
|
Release | - |
Please keep the discussion on the mailing list rather than commenting on the wiki (wiki discussions get unwieldy fast).
[This FLIP proposal is a joint work of Ran Jinhao, Dong Lin and Xuannan Su ]
As shown in the figure below, we might have a job that pre-processes records from a bounded source (i.e. inputA) using an operator (i.e. operatorA) which only emits results after its input has ended. Then operatorB needs to join records emitted by operatorA with records from an unbounded source, and emit results with low processing latency in real-time.
Currently, supporting the above use-case requires all operators to be deployed at the start of the job. This approach wastes slot and memory resources because operatorB can not do any useful work until operatorA's input has ended. Even worse, operatorB might use a lot of disk space only to cache and spill records received from the unbounded source to its local disk while it is waiting for operatorA's output.
In this FLIP, we propose to optimize task deployment and resource utilization for the above use-case by allowing an operator to explicitly specify whether it only emits records after all its inputs have ended. JM will leverage this information to optimize job scheduling such that the partition type of the results emitted by this operator will be blocking, which effectively let Flink schedule and execute this operator in batch mode.
In addition, this FLIP also adds EndOfStreamTrigger in GlobalWindows. GlobalWindows with EndOfStreamTrigger can be used with the DataStream API to specify whether the window will end only at the end of its inputs. DataStream program (e.g. coGroup and aggregate) can take advantage of this information to significantly increase throughput and reduce resource utilization.
Objectives
The APIs provided in this FLIP achieves the following objectives / benefits:
1) Improve throughput. For use-case that involves a mixture of bounded and unbounded workloads (e.g. the use-case specified in the example), the user can start a single Flink job to address such use-case where the performance of the bounded workload can be optimal (e.g. similar/higher than the corresponding performance in batch mode).
2) Reduce resource usage. For use-case that involves an operatorB (with unbounded input) depending on the output of another operatorA, where operatorA only emits results at the end of its input, Flink will deploy operatorB after operatorA is finished. This approach reduces the unnecessary resource usage when operatorA is still processing its inputs.
3) Improve usability. For use-case that needs to invoke DataStream APIs (e.g. KeyedStream#window) with a window assigner that covers all input data, users can use the off-the-shelf GlobalWindows with EndOfStreamTrigger provided in this FLIP, instead of writing tens of lines of code to define this WindowAssigner subclass. Existing operators can also redefine operator attributes to achieve this objective.
Public Interfaces
1) Add the EndOfStreamTrigger in GlobalWindows which allows users of the DataStream API to create GlobalWindows with EndOfStreamTrigger to specify whether the computation (e.g. co-group, aggregate) should emit data only after end-of-input.
@PublicEvolving public class GlobalWindows extends WindowAssigner<Object, GlobalWindow> { ··· public static GlobalWindows create() { return new GlobalWindows(new NeverTrigger()); } public static GlobalWindows createWithEndOfStreamTrigger() { return new GlobalWindows(new EndOfStreamTrigger()); } }
2) Add OperatorAttributesBuilder and OperatorAttributes for operator developers to specify operator attributes that Flink runtime can use to optimize the job performance.
package org.apache.flink.streaming.api.operators; /** The builder class for {@link OperatorAttributes}. */ @Experimental public class OperatorAttributesBuilder { @Nullable private Boolean outputOnEOF = null; public OperatorAttributesBuilder() {...} public OperatorAttributesBuilder setOutputOnEOF(boolean outputOnEOF) {...} /** * If any operator attribute is null, we will log it at DEBUG level and use the following * default values. * - outputOnEOF defaults to false */ public OperatorAttributes build() {...} }
package org.apache.flink.streaming.api.operators; /** * OperatorAttributes element provides Job Manager with information that can be * used to optimize the job performance. */ @Experimental public class OperatorAttributes { /** * Returns true iff the operator can only emit records after inputs have reached EOF. * * <p>Here are the implications when it is true: * * <ul> * <li> The results of this operator as well as its chained operators have blocking partition type. * <li> This operator as well as its chained operators will be executed in batch mode. * </ul> */ public boolean isOutputOnEOF() {...} }
3) Add the getOperatorAttributes() API to the StreamOperator and StreamOperatorFactory interfaces.
public interface StreamOperator<OUT> extends CheckpointListener, KeyContext, Serializable { ... @Experimental default OperatorAttributes getOperatorAttributes() { return new OperatorAttributesBuilder().build(); } } public interface StreamOperatorFactory<OUT> extends Serializable { ... @Experimental default OperatorAttributes getOperatorAttributes() { return new OperatorAttributesBuilder().build(); } }
Proposed Changes
1) Add the APIs on Transformation interface to get the corresponding operator attributes.
@Internal public abstract class Transformation<T> { public boolean isOutputOnEOF() { return false; } }
2) Update Transformation subclasses (e.g. OneInputTransformation and TwoInputTransformation) to override the newly added methods using the OperatorAttributes obtained from the underlying Operator.
3) Update JM to make use of the following operator attributes when compiling the Transformation graph into the JobGraph.
If a Transformation has isOutputOnEOF == true:
- The process of operator chain can still be done. After that, the results of its operator chain will be set blocking.
- This operator will be executed in batch mode.
More specifically, the checkpoint is disabled when these operators are running, such that these operators can do operations not compatible with checkpoints (e.g. sorting inputs). And operators should re-read the data from the upstream blocking edge or sources after failover.
4) All existing operators can redefine operator attributes,isOutputOnEOF == true, to use new features metioned in 3).
5) When DataStream API like DataStream#CoGroup is invoked with GlobalWindows andEndOfStreamTrigger as the window assigner and trigger, Flink should generate an operator with isOutputOnEOF = true to achieve higher throughput by using the optimizations in batch mode.
This FLIP can potentially benefit all DataStream APIs that take WindowAssigner as the parameter.
More specifically, the following DataStream API can benefit from using GlobalWindows with EndOfStreamTrigger.
- DataStream#windowAll
- KeyedStream#window
- CoGroupedStreams#Where#EqualTo#window
- JoinedStreams#Where#EqualTo#window
Benchmark results
To demonstrate our optimization improvements,we run each benchmark in different execution modes and configurations. Each result is measured by taking the average execution time across 5 runs with the given configuration.
jobmanager.memory.process.size: 6400m taskmanager.memory.process.size: 6912m
DataStream<Tuple2<Integer, Double>> source1 = env.fromCollection( new DataGenerator(dataNum), Types.TUPLE(Types.INT, Types.DOUBLE)); DataStream<Tuple2<Integer, Double>> source2 = env.fromCollection( new DataGenerator(dataNum), Types.TUPLE(Types.INT, Types.DOUBLE)); source1.coGroup(source2) .where(tuple -> tuple.f0) .equalTo(tuple -> tuple.f0) .window(GlobalWindows.createWithEndOfStreamTrigger()) .apply(new CustomCoGroupFunction()) .addSink(...); private static class CustomCoGroupFunction extends RichCoGroupFunction<Tuple2<Integer, Double>, Tuple2<Integer, Double>, Integer> { @Override public void coGroup( Iterable<Tuple2<Integer, Double>> iterableA, Iterable<Tuple2<Integer, Double>> iterableB, Collector<Integer> collector) { collector.collect(1); } }
The following result shows the throughput (records/sec) when the benchmark is executed in streaming mode, batch mode and optimized streaming mode after this PR.
The result shows that DataStream#coGroup in optimized streaming mode can be 22X as fast as streaming mode and 3X as fast as batch mode
STREAMING | BATCH | Optimized STREAMING |
---|---|---|
66 ± 1 (100%, 1202426 ms) | 491 ± 5 (743%, 162731 ms) | 1506 ± 10 (2281%, 53098 ms) |
2) Execute DataStream#Aggregate
DataStreamSource<Tuple2<Long, Double>> source = env.fromCollection( new DataGenerator(dataNum, keyNum), Types.TUPLE(Types.LONG, Types.DOUBLE)); source.keyBy(value -> value.f0) .window(GlobalWindows.createWithEndOfStreamTrigger()) .aggregate(new Aggregator()) .addSink(...); public static class Aggregator implements AggregateFunction<Tuple2<Long, Double>, Tuple2<Long, Double>, Double> { @Override public Tuple2<Long, Double> createAccumulator() { return new Tuple2<Long, Double>(0L, 0.0); } @Override public Tuple2<Long, Double> add(Tuple2<Long, Double> myData, Tuple2<Long, Double> accData) { accData.f1 = accData.f1 + myData.f1; return accData; } @Override public Double getResult(Tuple2<Long, Double> result) { return result.f1; } @Override public Tuple2<Long, Double> merge(Tuple2<Long, Double> acc1, Tuple2<Long, Double> acc2) { acc1.f1 = acc1.f1 + acc2.f1; return acc1; } }
The following result shows the throughput (records/sec) when the benchmark is executed in streaming mode, batch mode and optimized streaming mode after this PR.
STREAMING | BATCH | Optimized STREAMING |
---|---|---|
163 ± 0 (100%, 490478 ms) | 1561 ± 16 (957%, 51237 ms) | 1733 ± 9 (1063%, 46143 ms) |
3) Execute a program that needs to fully process data from a bounded source before processing data from another unbounded source.
The following program demonstrates the scenario described in the motivation section. The program needs to pre-processes records from a bounded source (Source1) using an operator (Process1) which only emits results after its input has ended. Then anther operator(Process2) needs to process records emitted by Process1 with records from an unbounded source, and emit results with low processing latency in real-time.
source1.keyBy(value -> value.f0) .window(GlobalWindows.createWithEndOfStreamTrigger()) .aggregate(new MyAggregator()).name("Process1") .connect(source2.keyBy(value -> value.f0)) .transform("Process2", Types.INT, new MyProcessOperator()) .addSink(...);
We can use this program to demonstrate that the program requires less slot resources. More specifically, suppose we configure the standalone cluster with taskmanager.numberOfTaskSlots = 2, and set the Source1,Process1, Source2 and Process2 in 4 different SlotSharingGroups, the program will fail to be deployed before this FLIP. And the program can be deployed successfully after this FLIP. This is because Source2
and Process2
can be deplyed after Source1
,Process1
finished and released their slots.
Additionally, we can use this program to demonstrate that it can achieve higher performance because Process2
will not need to keep buffer records emitted by Source2
in its memory while Process1
has not reached EOF. More specifically, the program can fail with OOM before this FLIP when the number of records in inputs is high. And the program can finish successfully without OOM after this FLIP.
Compatibility, Deprecation, and Migration Plan
The changes made in this FLIP are backward compatible.
Future Work
- It would be useful to add an ExecutionState (e.g. Finishing) to specify whether the task has reached EOF for all its inputs. This allows JM to deploy its downstream tasks and possibly apply hybrid shuffle to increase job throughput.
- Setting the upstream operators of EOF operator in batch mode will also increase the job throughput, and Hybrid shuffle mode can also be used in batch mode part to further improve the performance when there are sufficient slot resources.
- Developing frequently used EOF operator, such as aggregate/CoGroup/Join/..., can achieve higher throughput by using the optimizations currently done in batch mode. We can instantiate internal sorter in the operator, and flink will not invoke WindowAssigner#assignWindows or triggerContext#onElement for each input record to reduce overhead.