Also use that profile in create-release.sh
cc pwendell -- Note that this means that we need `knitr` and `roxygen` installed on the machines used for building the release. Let me know if you need help with that.
Author: Shivaram Venkataraman <shivaram@cs.berkeley.edu>
Closes#6567 from shivaram/SPARK-8027 and squashes the following commits:
8dc8ecf [Shivaram Venkataraman] Add maven profile to build R package docs Also use that profile in create-release.sh
(cherry picked from commit cae9306c4f)
Signed-off-by: Shivaram Venkataraman <shivaram@cs.berkeley.edu>
Switch to the official Pyrolite release from the one published under `org.spark-project`. Thanks irmen for making the releases on Maven Central. We didn't upgrade to 4.6 because we don't have enough time for QA. I excludes `serpent` from its dependencies because we don't use it in Spark.
~~~
[info] +-net.jpountz.lz4:lz4:1.3.0
[info] +-net.razorvine:pyrolite:4.4
[info] +-net.sf.py4j:py4j:0.8.2.1
~~~
davies
Author: Xiangrui Meng <meng@databricks.com>
Closes#6472 from mengxr/SPARK-7926 and squashes the following commits:
7b3c6bf [Xiangrui Meng] use the official Pyrolite release
(cherry picked from commit c45d58c143)
Signed-off-by: Xiangrui Meng <meng@databricks.com>
This patch introduces a new shuffle manager that enhances the existing sort-based shuffle with a new cache-friendly sort algorithm that operates directly on binary data. The goals of this patch are to lower memory usage and Java object overheads during shuffle and to speed up sorting. It also lays groundwork for follow-up patches that will enable end-to-end processing of serialized records.
The new shuffle manager, `UnsafeShuffleManager`, can be enabled by setting `spark.shuffle.manager=tungsten-sort` in SparkConf.
The new shuffle manager uses directly-managed memory to implement several performance optimizations for certain types of shuffles. In cases where the new performance optimizations cannot be applied, the new shuffle manager delegates to SortShuffleManager to handle those shuffles.
UnsafeShuffleManager's optimizations will apply when _all_ of the following conditions hold:
- The shuffle dependency specifies no aggregation or output ordering.
- The shuffle serializer supports relocation of serialized values (this is currently supported
by KryoSerializer and Spark SQL's custom serializers).
- The shuffle produces fewer than 16777216 output partitions.
- No individual record is larger than 128 MB when serialized.
In addition, extra spill-merging optimizations are automatically applied when the shuffle compression codec supports concatenation of serialized streams. This is currently supported by Spark's LZF serializer.
At a high-level, UnsafeShuffleManager's design is similar to Spark's existing SortShuffleManager. In sort-based shuffle, incoming records are sorted according to their target partition ids, then written to a single map output file. Reducers fetch contiguous regions of this file in order to read their portion of the map output. In cases where the map output data is too large to fit in memory, sorted subsets of the output can are spilled to disk and those on-disk files are merged to produce the final output file.
UnsafeShuffleManager optimizes this process in several ways:
- Its sort operates on serialized binary data rather than Java objects, which reduces memory consumption and GC overheads. This optimization requires the record serializer to have certain properties to allow serialized records to be re-ordered without requiring deserialization. See SPARK-4550, where this optimization was first proposed and implemented, for more details.
- It uses a specialized cache-efficient sorter (UnsafeShuffleExternalSorter) that sorts arrays of compressed record pointers and partition ids. By using only 8 bytes of space per record in the sorting array, this fits more of the array into cache.
- The spill merging procedure operates on blocks of serialized records that belong to the same partition and does not need to deserialize records during the merge.
- When the spill compression codec supports concatenation of compressed data, the spill merge simply concatenates the serialized and compressed spill partitions to produce the final output partition. This allows efficient data copying methods, like NIO's `transferTo`, to be used and avoids the need to allocate decompression or copying buffers during the merge.
The shuffle read path is unchanged.
This patch is similar to [SPARK-4550](http://issues.apache.org/jira/browse/SPARK-4550) / #4450 but uses a slightly different implementation. The `unsafe`-based implementation featured in this patch lays the groundwork for followup patches that will enable sorting to operate on serialized data pages that will be prepared by Spark SQL's new `unsafe` operators (such as the new aggregation operator introduced in #5725).
### Future work
There are several tasks that build upon this patch, which will be left to future work:
- [SPARK-7271](https://issues.apache.org/jira/browse/SPARK-7271) Redesign / extend the shuffle interfaces to accept binary data as input. The goal here is to let us bypass serialization steps in cases where the sort input is produced by an operator that operates directly on binary data.
- Extension / redesign of the `Serializer` API. We can add new methods which allow serializers to determine the size requirements for serializing objects and for serializing objects directly to a specified memory address (similar to how `UnsafeRowConverter` works in Spark SQL).
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/apache/spark/5868)
<!-- Reviewable:end -->
Author: Josh Rosen <joshrosen@databricks.com>
Closes#5868 from JoshRosen/unsafe-sort and squashes the following commits:
ef0a86e [Josh Rosen] Fix scalastyle errors
7610f2f [Josh Rosen] Add tests for proper cleanup of shuffle data.
d494ffe [Josh Rosen] Fix deserialization of JavaSerializer instances.
52a9981 [Josh Rosen] Fix some bugs in the address packing code.
51812a7 [Josh Rosen] Change shuffle manager sort name to tungsten-sort
4023fa4 [Josh Rosen] Add @Private annotation to some Java classes.
de40b9d [Josh Rosen] More comments to try to explain metrics code
df07699 [Josh Rosen] Attempt to clarify confusing metrics update code
5e189c6 [Josh Rosen] Track time spend closing / flushing files; split TimeTrackingOutputStream into separate file.
d5779c6 [Josh Rosen] Merge remote-tracking branch 'origin/master' into unsafe-sort
c2ce78e [Josh Rosen] Fix a missed usage of MAX_PARTITION_ID
e3b8855 [Josh Rosen] Cleanup in UnsafeShuffleWriter
4a2c785 [Josh Rosen] rename 'sort buffer' to 'pointer array'
6276168 [Josh Rosen] Remove ability to disable spilling in UnsafeShuffleExternalSorter.
57312c9 [Josh Rosen] Clarify fileBufferSize units
2d4e4f4 [Josh Rosen] Address some minor comments in UnsafeShuffleExternalSorter.
fdcac08 [Josh Rosen] Guard against overflow when expanding sort buffer.
85da63f [Josh Rosen] Cleanup in UnsafeShuffleSorterIterator.
0ad34da [Josh Rosen] Fix off-by-one in nextInt() call
56781a1 [Josh Rosen] Rename UnsafeShuffleSorter to UnsafeShuffleInMemorySorter
e995d1a [Josh Rosen] Introduce MAX_SHUFFLE_OUTPUT_PARTITIONS.
e58a6b4 [Josh Rosen] Add more tests for PackedRecordPointer encoding.
4f0b770 [Josh Rosen] Attempt to implement proper shuffle write metrics.
d4e6d89 [Josh Rosen] Update to bit shifting constants
69d5899 [Josh Rosen] Remove some unnecessary override vals
8531286 [Josh Rosen] Add tests that automatically trigger spills.
7c953f9 [Josh Rosen] Add test that covers UnsafeShuffleSortDataFormat.swap().
e1855e5 [Josh Rosen] Fix a handful of misc. IntelliJ inspections
39434f9 [Josh Rosen] Avoid integer multiplication overflow in getMemoryUsage (thanks FindBugs!)
1e3ad52 [Josh Rosen] Delete unused ByteBufferOutputStream class.
ea4f85f [Josh Rosen] Roll back an unnecessary change in Spillable.
ae538dc [Josh Rosen] Document UnsafeShuffleManager.
ec6d626 [Josh Rosen] Add notes on maximum # of supported shuffle partitions.
0d4d199 [Josh Rosen] Bump up shuffle.memoryFraction to make tests pass.
b3b1924 [Josh Rosen] Properly implement close() and flush() in DummySerializerInstance.
1ef56c7 [Josh Rosen] Revise compression codec support in merger; test cross product of configurations.
b57c17f [Josh Rosen] Disable some overly-verbose logs that rendered DEBUG useless.
f780fb1 [Josh Rosen] Add test demonstrating which compression codecs support concatenation.
4a01c45 [Josh Rosen] Remove unnecessary log message
27b18b0 [Josh Rosen] That for inserting records AT the max record size.
fcd9a3c [Josh Rosen] Add notes + tests for maximum record / page sizes.
9d1ee7c [Josh Rosen] Fix MiMa excludes for ShuffleWriter change
fd4bb9e [Josh Rosen] Use own ByteBufferOutputStream rather than Kryo's
67d25ba [Josh Rosen] Update Exchange operator's copying logic to account for new shuffle manager
8f5061a [Josh Rosen] Strengthen assertion to check partitioning
01afc74 [Josh Rosen] Actually read data in UnsafeShuffleWriterSuite
1929a74 [Josh Rosen] Update to reflect upstream ShuffleBlockManager -> ShuffleBlockResolver rename.
e8718dd [Josh Rosen] Merge remote-tracking branch 'origin/master' into unsafe-sort
9b7ebed [Josh Rosen] More defensive programming RE: cleaning up spill files and memory after errors
7cd013b [Josh Rosen] Begin refactoring to enable proper tests for spilling.
722849b [Josh Rosen] Add workaround for transferTo() bug in merging code; refactor tests.
9883e30 [Josh Rosen] Merge remote-tracking branch 'origin/master' into unsafe-sort
b95e642 [Josh Rosen] Refactor and document logic that decides when to spill.
1ce1300 [Josh Rosen] More minor cleanup
5e8cf75 [Josh Rosen] More minor cleanup
e67f1ea [Josh Rosen] Remove upper type bound in ShuffleWriter interface.
cfe0ec4 [Josh Rosen] Address a number of minor review comments:
8a6fe52 [Josh Rosen] Rename UnsafeShuffleSpillWriter to UnsafeShuffleExternalSorter
11feeb6 [Josh Rosen] Update TODOs related to shuffle write metrics.
b674412 [Josh Rosen] Merge remote-tracking branch 'origin/master' into unsafe-sort
aaea17b [Josh Rosen] Add comments to UnsafeShuffleSpillWriter.
4f70141 [Josh Rosen] Fix merging; now passes UnsafeShuffleSuite tests.
133c8c9 [Josh Rosen] WIP towards testing UnsafeShuffleWriter.
f480fb2 [Josh Rosen] WIP in mega-refactoring towards shuffle-specific sort.
57f1ec0 [Josh Rosen] WIP towards packed record pointers for use in optimized shuffle sort.
69232fd [Josh Rosen] Enable compressible address encoding for off-heap mode.
7ee918e [Josh Rosen] Re-order imports in tests
3aeaff7 [Josh Rosen] More refactoring and cleanup; begin cleaning iterator interfaces
3490512 [Josh Rosen] Misc. cleanup
f156a8f [Josh Rosen] Hacky metrics integration; refactor some interfaces.
2776aca [Josh Rosen] First passing test for ExternalSorter.
5e100b2 [Josh Rosen] Super-messy WIP on external sort
595923a [Josh Rosen] Remove some unused variables.
8958584 [Josh Rosen] Fix bug in calculating free space in current page.
f17fa8f [Josh Rosen] Add missing newline
c2fca17 [Josh Rosen] Small refactoring of SerializerPropertiesSuite to enable test re-use:
b8a09fe [Josh Rosen] Back out accidental log4j.properties change
bfc12d3 [Josh Rosen] Add tests for serializer relocation property.
240864c [Josh Rosen] Remove PrefixComputer and require prefix to be specified as part of insert()
1433b42 [Josh Rosen] Store record length as int instead of long.
026b497 [Josh Rosen] Re-use a buffer in UnsafeShuffleWriter
0748458 [Josh Rosen] Port UnsafeShuffleWriter to Java.
87e721b [Josh Rosen] Renaming and comments
d3cc310 [Josh Rosen] Flag that SparkSqlSerializer2 supports relocation
e2d96ca [Josh Rosen] Expand serializer API and use new function to help control when new UnsafeShuffle path is used.
e267cee [Josh Rosen] Fix compilation of UnsafeSorterSuite
9c6cf58 [Josh Rosen] Refactor to use DiskBlockObjectWriter.
253f13e [Josh Rosen] More cleanup
8e3ec20 [Josh Rosen] Begin code cleanup.
4d2f5e1 [Josh Rosen] WIP
3db12de [Josh Rosen] Minor simplification and sanity checks in UnsafeSorter
767d3ca [Josh Rosen] Fix invalid range in UnsafeSorter.
e900152 [Josh Rosen] Add test for empty iterator in UnsafeSorter
57a4ea0 [Josh Rosen] Make initialSize configurable in UnsafeSorter
abf7bfe [Josh Rosen] Add basic test case.
81d52c5 [Josh Rosen] WIP on UnsafeSorter
(cherry picked from commit 73bed408fb)
Signed-off-by: Reynold Xin <rxin@databricks.com>
The sbt part of the build is hacky; it basically tricks sbt
into generating the zip by using a generator, but returns
an empty list for the generated files so that nothing is
actually added to the assembly.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#6022 from vanzin/SPARK-7485 and squashes the following commits:
22c1e04 [Marcelo Vanzin] Remove unneeded code.
4893622 [Marcelo Vanzin] [SPARK-7485] [build] Remove pyspark files from assembly.
(cherry picked from commit 82e890fb19)
Signed-off-by: Andrew Or <andrew@databricks.com>
Exposes data available in the UI as json over http. Key points:
* new endpoints, handled independently of existing XyzPage classes. Root entrypoint is `JsonRootResource`
* Uses jersey + jackson for routing & converting POJOs into json
* tests against known results in `HistoryServerSuite`
* also fixes some minor issues w/ the UI -- synchronizing on access to `StorageListener` & `StorageStatusListener`, and fixing some inconsistencies w/ the way we handle retained jobs & stages.
Author: Imran Rashid <irashid@cloudera.com>
Closes#5940 from squito/SPARK-3454_better_test_files and squashes the following commits:
1a72ed6 [Imran Rashid] rats
85fdb3e [Imran Rashid] Merge branch 'no_php' into SPARK-3454
1fc65b0 [Imran Rashid] Revert "Revert "[SPARK-3454] separate json endpoints for data in the UI""
1276900 [Imran Rashid] get rid of giant event file, replace w/ smaller one; check both shuffle read & shuffle write
4e12013 [Imran Rashid] just use test case name for expectation file name
863ef64 [Imran Rashid] rename json files to avoid strange file names and not look like php
(cherry picked from commit c796be70f3)
Signed-off-by: Patrick Wendell <patrick@databricks.com>
Exposes data available in the UI as json over http. Key points:
* new endpoints, handled independently of existing XyzPage classes. Root entrypoint is `JsonRootResource`
* Uses jersey + jackson for routing & converting POJOs into json
* tests against known results in `HistoryServerSuite`
* also fixes some minor issues w/ the UI -- synchronizing on access to `StorageListener` & `StorageStatusListener`, and fixing some inconsistencies w/ the way we handle retained jobs & stages.
Author: Imran Rashid <irashid@cloudera.com>
Closes#4435 from squito/SPARK-3454 and squashes the following commits:
da1e35f [Imran Rashid] typos etc.
5e78b4f [Imran Rashid] fix rendering problems
5ae02ad [Imran Rashid] Merge branch 'master' into SPARK-3454
f016182 [Imran Rashid] change all constructors json-pojo class constructors to be private[spark] to protect us from mima-false-positives if we add fields
3347b72 [Imran Rashid] mark EnumUtil as @Private
ec140a2 [Imran Rashid] create @Private
cc1febf [Imran Rashid] add docs on the metrics-as-json api
cbaf287 [Imran Rashid] Merge branch 'master' into SPARK-3454
56db31e [Imran Rashid] update tests for mulit-attempt
7f3bc4e [Imran Rashid] Revert "add sbt-revolved plugin, to make it easier to start & stop http servers in sbt"
67008b4 [Imran Rashid] rats
9e51400 [Imran Rashid] style
c9bae1c [Imran Rashid] handle multiple attempts per app
b87cd63 [Imran Rashid] add sbt-revolved plugin, to make it easier to start & stop http servers in sbt
188762c [Imran Rashid] multi-attempt
2af11e5 [Imran Rashid] Merge branch 'master' into SPARK-3454
befff0c [Imran Rashid] review feedback
14ac3ed [Imran Rashid] jersey-core needs to be explicit; move version & scope to parent pom.xml
f90680e [Imran Rashid] Merge branch 'master' into SPARK-3454
dc8a7fe [Imran Rashid] style, fix errant comments
acb7ef6 [Imran Rashid] fix indentation
7bf1811 [Imran Rashid] move MetricHelper so mima doesnt think its exposed; comments
9d889d6 [Imran Rashid] undo some unnecessary changes
f48a7b0 [Imran Rashid] docs
52bbae8 [Imran Rashid] StorageListener & StorageStatusListener needs to synchronize internally to be thread-safe
31c79ce [Imran Rashid] asm no longer needed for SPARK_PREPEND_CLASSES
b2f8b91 [Imran Rashid] @DeveloperApi
2e19be2 [Imran Rashid] lazily convert ApplicationInfo to avoid memory overhead
ba3d9d2 [Imran Rashid] upper case enums
39ac29c [Imran Rashid] move EnumUtil
d2bde77 [Imran Rashid] update error handling & scoping
4a234d3 [Imran Rashid] avoid jersey-media-json-jackson b/c of potential version conflicts
a157a2f [Imran Rashid] style
7bd4d15 [Imran Rashid] delete security test, since it doesnt do anything
a325563 [Imran Rashid] style
a9c5cf1 [Imran Rashid] undo changes superceeded by master
0c6f968 [Imran Rashid] update deps
1ed0d07 [Imran Rashid] Merge branch 'master' into SPARK-3454
4c92af6 [Imran Rashid] style
f2e63ad [Imran Rashid] Merge branch 'master' into SPARK-3454
c22b11f [Imran Rashid] fix compile error
9ea682c [Imran Rashid] go back to good ol' java enums
cf86175 [Imran Rashid] style
d493b38 [Imran Rashid] Merge branch 'master' into SPARK-3454
f05ae89 [Imran Rashid] add in ExecutorSummaryInfo for MiMa :(
101a698 [Imran Rashid] style
d2ef58d [Imran Rashid] revert changes that had HistoryServer refresh the application listing more often
b136e39b [Imran Rashid] Revert "add sbt-revolved plugin, to make it easier to start & stop http servers in sbt"
e031719 [Imran Rashid] fixes from review
1f53a66 [Imran Rashid] style
b4a7863 [Imran Rashid] fix compile error
2c8b7ee [Imran Rashid] rats
1578a4a [Imran Rashid] doc
674f8dc [Imran Rashid] more explicit about total numbers of jobs & stages vs. number retained
9922be0 [Imran Rashid] Merge branch 'master' into stage_distributions
f5a5196 [Imran Rashid] undo removal of renderJson from MasterPage, since there is no substitute yet
db61211 [Imran Rashid] get JobProgressListener directly from UI
fdfc181 [Imran Rashid] stage/taskList
63eb4a6 [Imran Rashid] tests for taskSummary
ad27de8 [Imran Rashid] error handling on quantile values
b2efcaf [Imran Rashid] cleanup, combine stage-related paths into one resource
aaba896 [Imran Rashid] wire up task summary
a4b1397 [Imran Rashid] stage metric distributions
e48ba32 [Imran Rashid] rename
eaf3bbb [Imran Rashid] style
25cd894 [Imran Rashid] if only given day, assume GMT
51eaedb [Imran Rashid] more visibility fixes
9f28b7e [Imran Rashid] ack, more cleanup
99764e1 [Imran Rashid] Merge branch 'SPARK-3454_w_jersey' into SPARK-3454
a61a43c [Imran Rashid] oops, remove accidental checkin
a066055 [Imran Rashid] set visibility on a lot of classes
1f361c8 [Imran Rashid] update rat-excludes
0be5120 [Imran Rashid] Merge branch 'master' into SPARK-3454_w_jersey
2382bef [Imran Rashid] switch to using new "enum"
fef6605 [Imran Rashid] some utils for working w/ new "enum" format
dbfc7bf [Imran Rashid] style
b86bcb0 [Imran Rashid] update test to look at one stage attempt
5f9df24 [Imran Rashid] style
7fd156a [Imran Rashid] refactor jsonDiff to avoid code duplication
73f1378 [Imran Rashid] test json; also add test cases for cleaned stages & jobs
97d411f [Imran Rashid] json endpoint for one job
0c96147 [Imran Rashid] better error msgs for bad stageId vs bad attemptId
dddbd29 [Imran Rashid] stages have attempt; jobs are sorted; resource for all attempts for one stage
190c17a [Imran Rashid] StagePage should distinguish no task data, from unknown stage
84cd497 [Imran Rashid] AllJobsPage should still report correct completed & failed job count, even if some have been cleaned, to make it consistent w/ AllStagesPage
36e4062 [Imran Rashid] SparkUI needs to know about startTime, so it can list its own applicationInfo
b4c75ed [Imran Rashid] fix merge conflicts; need to widen visibility in a few cases
e91750a [Imran Rashid] Merge branch 'master' into SPARK-3454_w_jersey
56d2fc7 [Imran Rashid] jersey needs asm for SPARK_PREPEND_CLASSES to work
f7df095 [Imran Rashid] add test for accumulables, and discover that I need update after all
9c0c125 [Imran Rashid] add accumulableInfo
00e9cc5 [Imran Rashid] more style
3377e61 [Imran Rashid] scaladoc
d05f7a9 [Imran Rashid] dont use case classes for status api POJOs, since they have binary compatibility issues
654cecf [Imran Rashid] move all the status api POJOs to one file
b86e2b0 [Imran Rashid] style
18a8c45 [Imran Rashid] Merge branch 'master' into SPARK-3454_w_jersey
5598f19 [Imran Rashid] delete some unnecessary code, more to go
56edce0 [Imran Rashid] style
017c755 [Imran Rashid] add in metrics now available
1b78cb7 [Imran Rashid] fix some import ordering
0dc3ea7 [Imran Rashid] if app isnt found, reload apps from FS before giving up
c7d884f [Imran Rashid] fix merge conflicts
0c12b50 [Imran Rashid] Merge branch 'master' into SPARK-3454_w_jersey
b6a96a8 [Imran Rashid] compare json by AST, not string
cd37845 [Imran Rashid] switch to using java.util.Dates for times
a4ab5aa [Imran Rashid] add in explicit dependency on jersey 1.9 -- maven wasn't happy before this
4fdc39f [Imran Rashid] refactor case insensitive enum parsing
cba1ef6 [Imran Rashid] add security (maybe?) for metrics json
f0264a7 [Imran Rashid] switch to using jersey for metrics json
bceb3a9 [Imran Rashid] set http response code on error, some testing
e0356b6 [Imran Rashid] put new test expectation files in rat excludes (is this OK?)
b252e7a [Imran Rashid] small cleanup of accidental changes
d1a8c92 [Imran Rashid] add sbt-revolved plugin, to make it easier to start & stop http servers in sbt
4b398d0 [Imran Rashid] expose UI data as json in new endpoints
(cherry picked from commit d49735800d)
Signed-off-by: Imran Rashid <irashid@cloudera.com>
Without this, any dependency that pulls ivy transitively may override
the version and potentially cause issue. In my machine, the hive tests
were pulling an old version of ivy, and subsequently failing with a
"NoSuchMethodError".
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#5893 from vanzin/ivy-dep-fix and squashes the following commits:
ea2112d [Marcelo Vanzin] [minor] [build] Declare ivy dependency in root pom.
(cherry picked from commit c5790a2f77)
Signed-off-by: Sean Owen <sowen@cloudera.com>
This PR upgrades Pyrolite to 4.4, which contains the bug fix for SPARK-3524 and some other performance improvements (e.g., SPARK-6288). The artifact is still under `org.spark-project` on Maven Central since there is no official release published there.
Author: Xiangrui Meng <meng@databricks.com>
Closes#5850 from mengxr/SPARK-7314 and squashes the following commits:
2ed4a95 [Xiangrui Meng] Merge remote-tracking branch 'apache/master' into SPARK-7314
da3c2dd [Xiangrui Meng] remove my repo
fe7e29b [Xiangrui Meng] switch to maven central
6ddac0e [Xiangrui Meng] reverse the machine code for float/double
d2d5b5b [Xiangrui Meng] change back to 4.4
7824a9c [Xiangrui Meng] use Pyrolite 3.1
cc3903a [Xiangrui Meng] upgrade Pyrolite to 4.4-0 for testing
(cherry picked from commit e9b16e67c6)
Signed-off-by: Xiangrui Meng <meng@databricks.com>
This patch adds managed-memory-based aggregation to Spark SQL / DataFrames. Instead of working with Java objects, this new aggregation path uses `sun.misc.Unsafe` to manipulate raw memory. This reduces the memory footprint for aggregations, resulting in fewer spills, OutOfMemoryErrors, and garbage collection pauses. As a result, this allows for higher memory utilization. It can also result in better cache locality since objects will be stored closer together in memory.
This feature can be eanbled by setting `spark.sql.unsafe.enabled=true`. For now, this feature is only supported when codegen is enabled and only supports aggregations for which the grouping columns are primitive numeric types or strings and aggregated values are numeric.
### Managing memory with sun.misc.Unsafe
This patch supports both on- and off-heap managed memory.
- In on-heap mode, memory addresses are identified by the combination of a base Object and an offset within that object.
- In off-heap mode, memory is addressed directly with 64-bit long addresses.
To support both modes, functions that manipulate memory accept both `baseObject` and `baseOffset` fields. In off-heap mode, we simply pass `null` as `baseObject`.
We allocate memory in large chunks, so memory fragmentation and allocation speed are not significant bottlenecks.
By default, we use on-heap mode. To enable off-heap mode, set `spark.unsafe.offHeap=true`.
To track allocated memory, this patch extends `SparkEnv` with an `ExecutorMemoryManager` and supplies each `TaskContext` with a `TaskMemoryManager`. These classes work together to track allocations and detect memory leaks.
### Compact tuple format
This patch introduces `UnsafeRow`, a compact row layout. In this format, each tuple has three parts: a null bit set, fixed length values, and variable-length values:
- Rows are always 8-byte word aligned (so their sizes will always be a multiple of 8 bytes)
- The bit set is used for null tracking:
- Position _i_ is set if and only if field _i_ is null
- The bit set is aligned to an 8-byte word boundary.
- Every field appears as an 8-byte word in the fixed-length values part:
- If a field is null, we zero out the values.
- If a field is variable-length, the word stores a relative offset (w.r.t. the base of the tuple) that points to the beginning of the field's data in the variable-length part.
- Each variable-length data type can have its own encoding:
- For strings, the first word stores the length of the string and is followed by UTF-8 encoded bytes. If necessary, the end of the string is padded with empty bytes in order to ensure word-alignment.
For example, a tuple that consists 3 fields of type (int, string, string), with value (null, “data”, “bricks”) would look like this:
This format allows us to compare tuples for equality by directly comparing their raw bytes. This also enables fast hashing of tuples.
### Hash map for performing aggregations
This patch introduces `UnsafeFixedWidthAggregationMap`, a hash map for performing aggregations where the aggregation result columns are fixed-with. This map's keys and values are `Row` objects. `UnsafeFixedWidthAggregationMap` is implemented on top of `BytesToBytesMap`, an append-only map which supports byte-array keys and values.
`BytesToBytesMap` stores pointers to key and value tuples. For each record with a new key, we copy the key and create the aggregation value buffer for that key and put them in a buffer. The hash table then simply stores pointers to the key and value. For each record with an existing key, we simply run the aggregation function to update the values in place.
This map is implemented using open hashing with triangular sequence probing. Each entry stores two words in a long array: the first word stores the address of the key and the second word stores the relative offset from the key tuple to the value tuple, as well as the key's 32-bit hashcode. By storing the full hashcode, we reduce the number of equality checks that need to be performed to handle position collisions ()since the chance of hashcode collision is much lower than position collision).
`UnsafeFixedWidthAggregationMap` allows regular Spark SQL `Row` objects to be used when probing the map. Internally, it encodes these rows into `UnsafeRow` format using `UnsafeRowConverter`. This conversion has a small overhead that can be eliminated in the future once we use UnsafeRows in other operators.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/apache/spark/5725)
<!-- Reviewable:end -->
Author: Josh Rosen <joshrosen@databricks.com>
Closes#5725 from JoshRosen/unsafe and squashes the following commits:
eeee512 [Josh Rosen] Add converters for Null, Boolean, Byte, and Short columns.
81f34f8 [Josh Rosen] Follow 'place children last' convention for GeneratedAggregate
1bc36cc [Josh Rosen] Refactor UnsafeRowConverter to avoid unnecessary boxing.
017b2dc [Josh Rosen] Remove BytesToBytesMap.finalize()
50e9671 [Josh Rosen] Throw memory leak warning even in case of error; add warning about code duplication
70a39e4 [Josh Rosen] Split MemoryManager into ExecutorMemoryManager and TaskMemoryManager:
6e4b192 [Josh Rosen] Remove an unused method from ByteArrayMethods.
de5e001 [Josh Rosen] Fix debug vs. trace in logging message.
a19e066 [Josh Rosen] Rename unsafe Java test suites to match Scala test naming convention.
78a5b84 [Josh Rosen] Add logging to MemoryManager
ce3c565 [Josh Rosen] More comments, formatting, and code cleanup.
529e571 [Josh Rosen] Measure timeSpentResizing in nanoseconds instead of milliseconds.
3ca84b2 [Josh Rosen] Only zero the used portion of groupingKeyConversionScratchSpace
162caf7 [Josh Rosen] Fix test compilation
b45f070 [Josh Rosen] Don't redundantly store the offset from key to value, since we can compute this from the key size.
a8e4a3f [Josh Rosen] Introduce MemoryManager interface; add to SparkEnv.
0925847 [Josh Rosen] Disable MiMa checks for new unsafe module
cde4132 [Josh Rosen] Add missing pom.xml
9c19fc0 [Josh Rosen] Add configuration options for heap vs. offheap
6ffdaa1 [Josh Rosen] Null handling improvements in UnsafeRow.
31eaabc [Josh Rosen] Lots of TODO and doc cleanup.
a95291e [Josh Rosen] Cleanups to string handling code
afe8dca [Josh Rosen] Some Javadoc cleanup
f3dcbfe [Josh Rosen] More mod replacement
854201a [Josh Rosen] Import and comment cleanup
06e929d [Josh Rosen] More warning cleanup
ef6b3d3 [Josh Rosen] Fix a bunch of FindBugs and IntelliJ inspections
29a7575 [Josh Rosen] Remove debug logging
49aed30 [Josh Rosen] More long -> int conversion.
b26f1d3 [Josh Rosen] Fix bug in murmur hash implementation.
765243d [Josh Rosen] Enable optional performance metrics for hash map.
23a440a [Josh Rosen] Bump up default hash map size
628f936 [Josh Rosen] Use ints intead of longs for indexing.
92d5a06 [Josh Rosen] Address a number of minor code review comments.
1f4b716 [Josh Rosen] Merge Unsafe code into the regular GeneratedAggregate, guarded by a configuration flag; integrate planner support and re-enable all tests.
d85eeff [Josh Rosen] Add basic sanity test for UnsafeFixedWidthAggregationMap
bade966 [Josh Rosen] Comment update (bumping to refresh GitHub cache...)
b3eaccd [Josh Rosen] Extract aggregation map into its own class.
d2bb986 [Josh Rosen] Update to implement new Row methods added upstream
58ac393 [Josh Rosen] Use UNSAFE allocator in GeneratedAggregate (TODO: make this configurable)
7df6008 [Josh Rosen] Optimizations related to zeroing out memory:
c1b3813 [Josh Rosen] Fix bug in UnsafeMemoryAllocator.free():
738fa33 [Josh Rosen] Add feature flag to guard UnsafeGeneratedAggregate
c55bf66 [Josh Rosen] Free buffer once iterator has been fully consumed.
62ab054 [Josh Rosen] Optimize for fact that get() is only called on String columns.
c7f0b56 [Josh Rosen] Reuse UnsafeRow pointer in UnsafeRowConverter
ae39694 [Josh Rosen] Add finalizer as "cleanup method of last resort"
c754ae1 [Josh Rosen] Now that the store*() contract has been stregthened, we can remove an extra lookup
f764d13 [Josh Rosen] Simplify address + length calculation in Location.
079f1bf [Josh Rosen] Some clarification of the BytesToBytesMap.lookup() / set() contract.
1a483c5 [Josh Rosen] First version that passes some aggregation tests:
fc4c3a8 [Josh Rosen] Sketch how the converters will be used in UnsafeGeneratedAggregate
53ba9b7 [Josh Rosen] Start prototyping Java Row -> UnsafeRow converters
1ff814d [Josh Rosen] Add reminder to free memory on iterator completion
8a8f9df [Josh Rosen] Add skeleton for GeneratedAggregate integration.
5d55cef [Josh Rosen] Add skeleton for Row implementation.
f03e9c1 [Josh Rosen] Play around with Unsafe implementations of more string methods.
ab68e08 [Josh Rosen] Begin merging the UTF8String implementations.
480a74a [Josh Rosen] Initial import of code from Databricks unsafe utils repo.
Update Maven build plugin versions and centralize plugin version management
Author: Sean Owen <sowen@cloudera.com>
Closes#5720 from srowen/SPARK-7168 and squashes the following commits:
98a8947 [Sean Owen] Make install, deploy plugin versions explicit
4ecf3b2 [Sean Owen] Update Maven build plugin versions and centralize plugin version management
