Skip to content

Spark Configuration Properties


Controls whether to enable ContextCleaner

Default: true

Address of the driver (endpoints)

Default: Utils.localCanonicalHostName


Port of the driver (endpoints)

Default: 0


Number of CPU cores for Executor

Default: 1

Default: (undefined)


The files to be added to a Spark application (that can be defined directly as a configuration property or indirectly using --files option of spark-submit script)

Default: (empty)

Used when:


Default: (empty)


Default: false


Master URL of the cluster manager to connect the Spark application to


Controls whether to compress RDD partitions when stored serialized

Default: false


Controls whether to compress RDD partitions when stored serialized

Default: false


Controls whether to compress shuffle output when stored

Default: true


If enabled (with spark.shuffle.useOldFetchProtocol disabled and spark.shuffle.service.enabled enabled), shuffle blocks requested from those block managers which are running on the same host are read from the disk directly instead of being fetched as remote blocks over the network.

Default: true

The max number of executors for which the local dirs are stored. This size is both applied for the driver and both for the executors side to avoid having an unbounded store. This cache will be used to avoid the network in case of fetching disk persisted RDD blocks or shuffle blocks (when spark.shuffle.readHostLocalDisk is set) from the same host.

Default: 1000


Whether to use the old protocol while doing the shuffle block fetching. It is only enabled while we need the compatibility in the scenario of new Spark version job fetching shuffle blocks from old version external shuffle service.

Default: false

Default: RandomBlockReplicationPolicy


Minimum ratio of (registered resources / total expected resources) before submitting tasks

Default: (undefined)


The number of CPU cores to schedule (allocate) to a task

Default: 1

Used when:


Revive Interval - time (in millis) between resource offers revives

Default: 1s

Used when:

  • DriverEndpoint is requested to onStart


Maximum allowed message size for RPC communication (in MB unless specified)

Default: 128

Must be below 2047MB (Int.MaxValue / 1024 / 1024)

Used when:


Maximum size of a task result (in bytes) to be sent to the driver as a DirectTaskResult

Default: 1048576B (1L << 20)

Used when:


Maximum size of task results (in bytes)

Default: 1g

Used when:

(in millis)

Default: 60s

(in millis)


Network timeout (in seconds) to use for RPC remote endpoint lookup

Default: 120s

(in millis)



How many attempts to register a BlockManager with External Shuffle Service

Default: 3

Used when BlockManager is requested to register with External Shuffle Server


Controls whether DiskBlockObjectWriter should force outstanding writes to disk while committing a single atomic block (i.e. all operating system buffers should synchronize with the disk to ensure that all changes to a file are in fact recorded in the storage)

Default: false

Used when BlockManager is requested for a DiskBlockObjectWriter


(internal) Minimum number of partitions (threshold) for MapStatus utility to prefer a HighlyCompressedMapStatus (over CompressedMapStatus) (for ShuffleWriters).

Default: 2000

Must be a positive integer (above 0)


Controls whether to use Java FileChannels (Java NIO) for copying data between two Java FileInputStreams to improve copy performance

Default: true

Used when BypassMergeSortShuffleWriter and UnsafeShuffleWriter are created


Number of subdirectories inside each path listed in spark.local.dir for hashing block files into.

Default: 64

Used by BlockManager and DiskBlockManager


A comma-separated list of directories that are used as a temporary storage for "scratch" space (incl. map output files and RDDs that get stored on disk). This should be on a fast, local disk in your system.

Default: /tmp


Controls whether to enable ConsoleProgressBar and show the progress bar in the console

Default: false


Default: false


Initial threshold for the size of an in-memory collection

Default: 5MB

Used by Spillable


A fully-qualified class name or the alias of the ShuffleManager in a Spark application

Default: sort

Supported aliases:

  • sort
  • tungsten-sort

Used when SparkEnv object is requested to create a "base" SparkEnv for a driver or an executor


(internal) The maximum number of elements in memory before forcing the shuffle sorter to spill.

Default: Integer.MAX_VALUE

The default value is to never force the sorter to spill, until Spark reaches some limitations, like the max page size limitation for the pointer array in the sorter.

Used when:

  • ShuffleExternalSorter is created
  • Spillable is created
  • Spark SQL's SortBasedAggregator is requested for an UnsafeKVExternalSorter
  • Spark SQL's ObjectAggregationMap is requested to dumpToExternalSorter
  • Spark SQL's UnsafeExternalRowSorter is created
  • Spark SQL's UnsafeFixedWidthAggregationMap is requested for an UnsafeKVExternalSorter


Size of the in-memory buffer for each shuffle file output stream, in KiB unless otherwise specified. These buffers reduce the number of disk seeks and system calls made in creating intermediate shuffle files.

Default: 32k

Must be greater than 0 and less than or equal to 2097151 ((Integer.MAX_VALUE - 15) / 1024)

Used when the following are created:


A comma-separated list of class names implementing org.apache.spark.api.plugin.SparkPlugin to load into a Spark application.

Default: (empty)

Since: 3.0.0

Set when SparkContext is created



Unique identifier of a Spark application that Spark uses to uniquely identify metric sources.

Default: TaskScheduler.applicationId()

Set when SparkContext is created


A comma-separated list of fully-qualified class names of SparkListeners (to be registered when SparkContext is created)

Default: (empty)

== [[properties]] Properties

[cols="1m,1",options="header",width="100%"] |=== | Name | Description

| spark.blockManager.port a| [[spark.blockManager.port]][[BLOCK_MANAGER_PORT]] Port to use for block managers to listen on when a more specific setting is not provided (i.e. <> for the driver).

Default: 0

In Spark on Kubernetes the default port is 7079

| spark.default.parallelism a| [[spark.default.parallelism]] Number of partitions to use for[HashPartitioner]

spark.default.parallelism corresponds to[default parallelism] of a scheduler backend and is as follows:

  • The number of threads for local/[LocalSchedulerBackend].
  • the number of CPU cores in[Spark on Mesos] and defaults to 8.
  • Maximum of totalCoreCount and 2 in[CoarseGrainedSchedulerBackend].

| spark.driver.blockManager.port a| [[spark.driver.blockManager.port]][[DRIVER_BLOCK_MANAGER_PORT]] Port the[block manager] on the driver listens on

Default: <>

| spark.executor.extraClassPath a| [[spark.executor.extraClassPath]][[EXECUTOR_CLASS_PATH]] User-defined class path for executors, i.e. URLs representing user-defined class path entries that are added to an executor's class path. URLs are separated by system-dependent path separator, i.e. : on Unix-like systems and ; on Microsoft Windows.

Default: (empty)

Used when:

  • Spark Standalone's StandaloneSchedulerBackend is requested to[start] (and creates a command for[])

  • Spark local's LocalSchedulerBackend is requested for the[user-defined class path for executors]

  • Spark on Mesos' MesosCoarseGrainedSchedulerBackend is requested to[create a command for CoarseGrainedExecutorBackend]

  • Spark on Mesos' MesosFineGrainedSchedulerBackend is requested to create a command for MesosExecutorBackend

  • Spark on Kubernetes' BasicExecutorFeatureStep is requested to configurePod

  • Spark on YARN's ExecutorRunnable is requested to[prepareEnvironment] (for CoarseGrainedExecutorBackend)

| spark.executor.extraJavaOptions a| [[spark.executor.extraJavaOptions]] Extra Java options of an[]

Used when Spark on YARN's ExecutorRunnable is requested to[prepare the command to launch CoarseGrainedExecutorBackend in a YARN container]

| spark.executor.extraLibraryPath a| [[spark.executor.extraLibraryPath]] Extra library paths separated by system-dependent path separator, i.e. : on Unix/MacOS systems and ; on Microsoft Windows

Used when Spark on YARN's ExecutorRunnable is requested to[prepare the command to launch CoarseGrainedExecutorBackend in a YARN container]

| spark.executor.uri a| [[spark.executor.uri]] Equivalent to SPARK_EXECUTOR_URI

| spark.executor.logs.rolling.time.interval a| [[spark.executor.logs.rolling.time.interval]]

| spark.executor.logs.rolling.strategy a| [[spark.executor.logs.rolling.strategy]]

| spark.executor.logs.rolling.maxRetainedFiles a| [[spark.executor.logs.rolling.maxRetainedFiles]]

| spark.executor.logs.rolling.maxSize a| [[spark.executor.logs.rolling.maxSize]]

| spark.executor.heartbeatInterval a| [[spark.executor.heartbeatInterval]] Interval after which an[] reports heartbeat and metrics for active tasks to the driver

Default: 10s

Refer to[Sending heartbeats and partial metrics for active tasks]

| spark.executor.heartbeat.maxFailures a| [[spark.executor.heartbeat.maxFailures]] Number of times an[] will try to send heartbeats to the driver before it gives up and exits (with exit code 56).

Default: 60

NOTE: Introduced in[SPARK-13522 Executor should kill itself when it's unable to heartbeat to the driver more than N times].

| spark.executor.instances a| [[spark.executor.instances]] Number of[] in use

Default: 0

| spark.executor.userClassPathFirst a| [[spark.executor.userClassPathFirst]] Flag to control whether to load classes in user jars before those in Spark jars

Default: false

| spark.executor.memory a| [[spark.executor.memory]] Amount of memory to use for an[]

Default: 1g

Equivalent to[SPARK_EXECUTOR_MEMORY] environment variable.

Refer to[Executor Memory -- spark.executor.memory or SPARK_EXECUTOR_MEMORY settings]

| spark.executor.port a| [[spark.executor.port]]

| spark.launcher.port a| [[spark.launcher.port]]

| spark.launcher.secret a| [[spark.launcher.secret]]

| spark.locality.wait a| [[spark.locality.wait]] For locality-aware delay scheduling for PROCESS_LOCAL, NODE_LOCAL, and RACK_LOCAL[TaskLocalities] when locality-specific setting is not set.

Default: 3s

| spark.locality.wait.node a| [[spark.locality.wait.node]] Scheduling delay for NODE_LOCAL[TaskLocality]

Default: The value of <>

| spark.locality.wait.process a| [[spark.locality.wait.process]] Scheduling delay for PROCESS_LOCAL[TaskLocality]

Default: The value of <>

| spark.locality.wait.rack a| [[spark.locality.wait.rack]] Scheduling delay for RACK_LOCAL[TaskLocality]

Default: The value of <>

| spark.logging.exceptionPrintInterval a| [[spark.logging.exceptionPrintInterval]] How frequently to reprint duplicate exceptions in full (in millis).

Default: 10000

| spark.scheduler.allocation.file a| [[spark.scheduler.allocation.file]] Path to the configuration file of <>

Default: fairscheduler.xml (on a Spark application's class path)

| spark.scheduler.executorTaskBlacklistTime a| [[spark.scheduler.executorTaskBlacklistTime]] How long to wait before a task can be re-launched on the executor where it once failed. It is to prevent repeated task failures due to executor failures.

Default: 0L

| spark.scheduler.mode a| [[spark.scheduler.mode]][[SCHEDULER_MODE_PROPERTY]] Scheduling Mode of the[TaskSchedulerImpl], i.e. case-insensitive name of the[scheduling mode] that TaskSchedulerImpl uses to choose between the <> for task scheduling (of tasks of jobs submitted for execution to the same SparkContext)

Default: FIFO

Supported values:

  • FAIR for fair sharing (of cluster resources)
  • FIFO (default) for queueing jobs one after another

Task scheduling is an algorithm that is used to assign cluster resources (CPU cores and memory) to tasks (that are part of jobs with one or more stages). Fair sharing allows for executing tasks of different jobs at the same time (that were all submitted to the same SparkContext). In FIFO scheduling mode a single SparkContext can submit a single job for execution only (regardless of how many cluster resources the job really use which could lead to a inefficient utilization of cluster resources and a longer execution of the Spark application overall).

Scheduling mode is particularly useful in multi-tenant environments in which a single SparkContext could be shared across different users (to make a cluster resource utilization more efficient).

TIP: Use web UI to know the current scheduling mode (e.g. <> tab as part of Spark Properties and <> tab as Scheduling Mode).

| spark.starvation.timeout a| [[spark.starvation.timeout]] Threshold above which Spark warns a user that an initial TaskSet may be starved

Default: 15s

| a| [[]]

| spark.task.maxFailures a| [[spark.task.maxFailures]] The number of individual task failures before giving up on the entire[TaskSet] and the job afterwards


  • 1 in[local]
  • maxFailures in[local-with-retries]
  • 4 in[cluster mode]

| spark.unsafe.exceptionOnMemoryLeak a| [[spark.unsafe.exceptionOnMemoryLeak]]


== [[spark.memory.offHeap.size]][[MEMORY_OFFHEAP_SIZE]] spark.memory.offHeap.size

Maximum memory (in bytes) for off-heap memory allocation.

Default: 0

This setting has no impact on heap memory usage, so if your executors' total memory consumption must fit within some hard limit then be sure to shrink your JVM heap size accordingly.

Must be set to a positive value when <> is enabled (true).

Must not be negative

== [[spark.memory.storageFraction]] spark.memory.storageFraction

Fraction of the memory to use for off-heap storage region.

Default: 0.5

== [[spark.memory.fraction]] spark.memory.fraction

spark.memory.fraction is the fraction of JVM heap space used for execution and storage.

Default: 0.6

== [[spark.memory.useLegacyMode]] spark.memory.useLegacyMode

Controls the type of[MemoryManager] to use. When enabled (i.e. true) it is the legacy[StaticMemoryManager] while[UnifiedMemoryManager] otherwise (i.e. false).

Default: false

== [[spark.memory.offHeap.enabled]] spark.memory.offHeap.enabled

spark.memory.offHeap.enabled controls whether Spark will attempt to use off-heap memory for certain operations (true) or not (false).

Default: false

Tracks whether Tungsten memory will be allocated on the JVM heap or off-heap (using sun.misc.Unsafe).

If enabled, <> has to be[greater than 0].

Used when MemoryManager is requested for[tungstenMemoryMode].

== [[spark.shuffle.spill.batchSize]] spark.shuffle.spill.batchSize

Size of object batches when reading or writing from serializers.

Default: 10000

Used by[ExternalAppendOnlyMap] and[ExternalSorter]

== [[spark.shuffle.mapOutput.dispatcher.numThreads]] spark.shuffle.mapOutput.dispatcher.numThreads

Default: 8

== [[spark.shuffle.mapOutput.minSizeForBroadcast]] spark.shuffle.mapOutput.minSizeForBroadcast

Size of serialized shuffle map output statuses when[MapOutputTrackerMaster] uses to determine whether to use a broadcast variable to send them to executors

Default: 512k

Must be below <> (to prevent sending an RPC message that is too large)

== [[spark.shuffle.reduceLocality.enabled]] spark.shuffle.reduceLocality.enabled

Enables locality preferences for reduce tasks

Default: true

When enabled (true), MapOutputTrackerMaster will[compute the preferred hosts] on which to run a given map output partition in a given shuffle, i.e. the nodes that the most outputs for that partition are on.

== [[spark.shuffle.sort.bypassMergeThreshold]] spark.shuffle.sort.bypassMergeThreshold

Maximum number of reduce partitions below which[SortShuffleManager] avoids merge-sorting data for no map-side aggregation

Default: 200

== [[spark.shuffle.sort.initialBufferSize]] spark.shuffle.sort.initialBufferSize

Initial buffer size for sorting


Used exclusively when UnsafeShuffleWriter is requested to[open] (and creates a[ShuffleExternalSorter])

== [[spark.shuffle.unsafe.file.output.buffer]] spark.shuffle.unsafe.file.output.buffer

The file system for this buffer size after each partition is written in unsafe shuffle writer. In KiB unless otherwise specified.

Default: 32k

Must be greater than 0 and less than or equal to 2097151 ((Integer.MAX_VALUE - 15) / 1024)

== [[spark.scheduler.maxRegisteredResourcesWaitingTime]] spark.scheduler.maxRegisteredResourcesWaitingTime

Time to wait for sufficient resources available

Default: 30s

== [[spark.shuffle.unsafe.fastMergeEnabled]] spark.shuffle.unsafe.fastMergeEnabled

Enables fast merge strategy for UnsafeShuffleWriter to[merge spill files].

Default: true

== [[spark.shuffle.spill.compress]] spark.shuffle.spill.compress

Controls whether to compress shuffle output temporarily spilled to disk.

Default: true

== [[spark.block.failures.beforeLocationRefresh]] spark.block.failures.beforeLocationRefresh

Default: 5

== [[]]

Controls whether to use IO encryption

Default: false

== [[spark.closure.serializer]] spark.closure.serializer[Serializer]

Default: org.apache.spark.serializer.JavaSerializer

== [[spark.serializer]] spark.serializer[Serializer]

Default: org.apache.spark.serializer.JavaSerializer

== [[]]

The default[CompressionCodec]

Default: lz4

== [[]]

The block size of the[LZ4CompressionCodec]

Default: 32k

== [[]]

The block size of the[SnappyCompressionCodec]

Default: 32k

== [[]]

The buffer size of the BufferedOutputStream of the[ZStdCompressionCodec]

Default: 32k

The buffer is used to avoid the overhead of excessive JNI calls while compressing or uncompressing small amount of data

== [[]]

The compression level of the[ZStdCompressionCodec]

Default: 1

The default level is the fastest of all with reasonably high compression ratio

== [[spark.buffer.size]] spark.buffer.size

Default: 65536

== [[spark.cleaner.referenceTracking.cleanCheckpoints]] spark.cleaner.referenceTracking.cleanCheckpoints

Enables cleaning checkpoint files when a checkpointed reference is out of scope

Default: false

== [[spark.cleaner.periodicGC.interval]] spark.cleaner.periodicGC.interval

Controls how often to trigger a garbage collection

Default: 30min

== [[spark.cleaner.referenceTracking.blocking]] spark.cleaner.referenceTracking.blocking

Controls whether the cleaning thread should block on cleanup tasks (other than shuffle, which is controlled by <>)

Default: true

== [[spark.cleaner.referenceTracking.blocking.shuffle]] spark.cleaner.referenceTracking.blocking.shuffle

Controls whether the cleaning thread should block on shuffle cleanup tasks.

Default: false

== [[spark.broadcast.blockSize]] spark.broadcast.blockSize

The size of a block (in kB unless the unit is specified)

Default: 4m

Used when[TorrentBroadcast stores brodcast blocks to BlockManager]

== [[spark.broadcast.compress]] spark.broadcast.compress

Controls broadcast compression

Default: true

Used when[TorrentBroadcast is created] and later when[it stores broadcast blocks to BlockManager]. Also in[SerializerManager].

== [[]]

Application Name

Default: (undefined)

== [[spark.rpc.lookupTimeout]] spark.rpc.lookupTimeout

Timeout to use for the[Default Endpoint Lookup Timeout]

Default: 120s

== [[spark.rpc.numRetries]] spark.rpc.numRetries

Number of attempts to send a message to and receive a response from a remote endpoint.

Default: 3

== [[spark.rpc.retry.wait]] spark.rpc.retry.wait

Time to wait between retries.

Default: 3s

== [[spark.rpc.askTimeout]] spark.rpc.askTimeout

Timeout for RPC ask calls

Default: 120s

== [[spark.speculation]] spark.speculation

Enables (true) or disables (false)[]

Default: false

== [[spark.speculation.interval]] spark.speculation.interval

The time interval to use before checking for speculative tasks in[].

Default: 100ms

== [[spark.speculation.multiplier]] spark.speculation.multiplier

Default: 1.5

== [[spark.speculation.quantile]] spark.speculation.quantile

The percentage of tasks that has not finished yet at which to start speculation in[].

Default: 0.75

== [[]]

Initial per-task memory size needed to store a block in memory.

Default: 1024 * 1024

Must be at most the[total amount of memory available for storage]

Used when MemoryStore is requested to[putIterator] and[putIteratorAsBytes]

Last update: 2021-01-11