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 between Xuannan Su  and Dong Lin  ]

Motivation

Currently, the default configuration for pipeline.object-reuse is set to false. This means that each record emitted by an operator undergoes deep-copying, where it is serialized and then deserialized before being passed to the next operator in the chain. However, in cases where the upstream and downstream operators do not store or access references to the input or output records, this deep-copy overhead becomes unnecessary and can result in suboptimal performance [1]. While users can manually set pipeline.object-reuse to true, they bear the responsibility of ensuring that no operator stores or accesses record references, which may introduce the risk of data corruption. Consequently, users are faced with the dilemma of choosing between suboptimal performance and the potential for data corruption.

In this FLIP, we propose adding APIs that operators can utilize to inform the Flink runtime whether it is safe to reuse the emitted records. This enhancement would enable Flink to maximize its performance using the default configuration, eliminating the need for users to provide additional information.

Public Interfaces

1) Add OperatorAttributesBuilder and OperatorAttributes for developers to specify operator attributes.

package org.apache.flink.streaming.api.operators;
 
/** The builder class for {@link OperatorAttributes}. */
@PublicEvolving
public class OperatorAttributesBuilder {
    ...
 
    public OperatorAttributesBuilder() {
      ...
      isObjectReuseCompliant = false;
    }

    public OperatorAttributesBuilder setIsObjectReuseCompliant(boolean isObjectReuseCompliant) {...}
 
    public OperatorAttributes build() {...}
}

package org.apache.flink.streaming.api.operators;
 
/**
 * OperatorAttributes element provides Job Manager with information that can be
 * used to optimize job performance.
 */
@PublicEvolving 
public class OperatorAttributes {   
    
    /**
     * Returns true if it is guaranteed that the operator will not store and access
     * reference to input or output records.
     */
	public boolean getIsObjectReuseCompliant() {...}
}

2) Add the getOperatorAttributes() API to the StreamOperator and StreamOperatorFactory interfaces.

@PublicEvolving
public interface StreamOperator<OUT> extends CheckpointListener, KeyContext, Serializable {
    ...
 
    default OperatorAttributes getOperatorAttributes() {
        return new OperatorAttributesBuilder().build();
    }
}
 
@PublicEvolving
public interface StreamOperatorFactory<OUT> extends Serializable {
    ...
 
    default OperatorAttributes getOperatorAttributes() {
        return new OperatorAttributesBuilder().build();
    }
}

3) Add the isObjectReuseCompliant method that returns false by default to the Function interface.

@Public
public interface Function extends java.io.Serializable {
    /**
     * Returns true if it is guaranteed that the function will not store and access
     * reference to the input or output records.
     */ 
    default boolean isObjectReuseCompliant() {
        return false;
    }
}

Proposed Changes

1) Update OperatorChain to take advantage of the isObjectReuseCompliant attribute.

Currently, the OperatorChain only considers the global execution configuration pipeline.object-reuse to determine whether to enable object-reuse for data transfer between chained operators.

We propose updating the OperatorChain as follows:

• If pipeline.object-reuse is set to true, records emitted by this operator will be re-used.

• Otherwise, if getIsObjectReuseCompliant() returns true and all the downstream operators are object-reuse compliant, records emitted by this operator will be re-used.

• Otherwise, records emitted by this operator will be deep-copied before being given to the next operator in the chain.

2) Update operators/functions defined in Flink codebase to override the getIsObjectReuseCompliant() as appropriate.

We will enumerate all operators/functions defined in Flink, identify those operators/functions which do not store and access reference to the input or output records, and override their getIsObjectReuseCompliant() method to return true.

After implementing the proposed change, in typical cases where jobs don't involve user-defined operators or functions, such as SQL jobs, Flink will only perform deep-copying of records when it is essential for ensuring correctness. For jobs that do incorporate user-defined operators or functions, operator/function developer will have the flexibility to avoid unnecessary deep-copy overhead by appropriately overriding the getIsObjectReuseCompliant() method. This empowers Flink to achieve optimal performance with default configuration.

Compatibility, Deprecation, and Migration Plan

The changes made in this FLIP are backward compatible. No deprecation plan is needed.

Migration plan: users are recommended to update their existing operator/function implementation to override the getIsObjectReuseCompliant() API if their operator/function does not store and access reference to the records that it has already emitted. We plan to write blogs to promote the new API.

Test Plan

The change will be covered with unit and integration tests.

Rejected Alternatives

Not yet.

[1] https://www.ververica.com/blog/a-journey-to-beating-flinks-sql-performance