@Internal public enum SubtaskStateMapper extends Enum<SubtaskStateMapper>
SubtaskStateMappernarrows down the subtasks that need to be read during rescaling to recover from a particular subtask when in-flight data has been stored in the checkpoint.
Mappings of old subtasks to new subtasks may be unique or non-unique. A unique assignment
means that a particular old subtask is only assigned to exactly one new subtask. Non-unique
assignments require filtering downstream. That means that the receiver side has to cross-verify
for a deserialized record if it truly belongs to the new subtask or not. Most
SubtaskStateMapper will only produce unique assignments and are thus optimal. Some rescaler,
RANGE, create a mixture of unique and non-unique mappings, where downstream
tasks need to filter on some mapped subtasks.
|Enum Constant and Description|
Extra state is redistributed to other subtasks without any specific guarantee (only that up- and downstream are matched).
Restores extra subtasks to the first subtask.
Replicates the state to all subtasks.
Remaps old ranges to new ranges.
Redistributes subtask state in a round robin fashion.
|Modifier and Type||Method and Description|
Returns a mapping new subtask index to all old subtask indexes.
Returns all old subtask indexes that need to be read to restore all buffers for the given new subtask index on rescale.
Returns true iff this mapper can potentially lead to ambiguous mappings where the different new subtasks map to the same old subtask.
Returns the enum constant of this type with the specified name.
Returns an array containing the constants of this enum type, in the order they are declared.
clone, compareTo, equals, finalize, getDeclaringClass, hashCode, name, ordinal, toString, valueOf
public static final SubtaskStateMapper ARBITRARY
public static final SubtaskStateMapper FIRST
public static final SubtaskStateMapper FULL
This strategy should only be used as a fallback.
public static final SubtaskStateMapper RANGE
old assignment: 0 -> [0;43); 1 -> [43;87); 2 -> [87;128)
new assignment: 0 -> [0;64]; 1 -> [64;128)
subtask 0 recovers data from old subtask 0 + 1 and subtask 1 recovers data from old subtask 1 + 2
For all downscale from n to [n-1 .. n/2], each new subtasks get exactly two old subtasks assigned.
For all upscale from n to [n+1 .. 2*n-1], most subtasks get two old subtasks assigned, except the two outermost.
Larger scale factors (
>2*n), will increase the number of old
subtasks accordingly. However, they will also create more unique assignment, where an old
subtask is exclusively assigned to a new subtask. Thus, the number of non-unique mappings is
upper bound by 2*n.
public static final SubtaskStateMapper ROUND_ROBIN
newIndex -> oldIndexes. The mapping is accessed by using
Bitset oldIndexes = mapping.get(newIndex).
oldParallelism < newParallelism, that mapping is trivial. For example if
oldParallelism = 6 and newParallelism = 10.
|New index||Old indexes|
oldParallelism > newParallelism, new indexes get multiple assignments by
wrapping around assignments in a round-robin fashion. For example if oldParallelism = 10 and
newParallelism = 4.
|New index||Old indexes|
|0||0, 4, 8|
|1||1, 5, 9|
public static final SubtaskStateMapper UNSUPPORTED
public static SubtaskStateMapper values()
for (SubtaskStateMapper c : SubtaskStateMapper.values()) System.out.println(c);
public static SubtaskStateMapper valueOf(String name)
public abstract int getOldSubtasks(int newSubtaskIndex, int oldNumberOfSubtasks, int newNumberOfSubtasks)
public RescaleMappings getNewToOldSubtasksMapping(int oldParallelism, int newParallelism)
public boolean isAmbiguous()
Copyright © 2014–2022 The Apache Software Foundation. All rights reserved.