apache datafusion-comet issues

apache / datafusion-comet

Apache DataFusion Comet Spark Accelerator

https://datafusion.apache.org/comet

Apache License 2.0

823 stars 163 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

minor: use defaults instead of hard-coding values

#1060 andygrove closed 2 weeks ago
0
Spark ColumnarToRowExec cannot pass CometBuffer safety check

#1059 viirya opened 2 weeks ago
1
doc: fix K8s links and doc

#1058 comphead closed 2 weeks ago
0
chore: Upgrade to DataFusion 43.0.0-rc1

#1057 andygrove closed 2 weeks ago
4
chore: Refactor UnaryExpr and MathExpr in protobuf

#1056 andygrove closed 2 weeks ago
0
fix: Avoid to call import and export Arrow array for native execution

#1055 kazuyukitanimura opened 2 weeks ago
15
WIP: Create separate instance of CometShuffleMemoryAllocator per plan

#1054 andygrove closed 2 weeks ago
1
minor: Refactor binary expr serde to reduce code duplication

#1053 andygrove closed 2 weeks ago
2
Initcap behaves differently in Spark and in DataFusion (also Comet)

#1052 Blizzara opened 2 weeks ago
1
show a mismatch for initcap between Spark and DataFusion

#1051 Blizzara opened 2 weeks ago
0
chore: Add safety check to CometBuffer

#1050 viirya opened 2 weeks ago
2
build: Add build package workflow

#1049 wangyum closed 1 week ago
0
Refactor Arrow Array and Schema allocation in ColumnReader and MetadataColumnReader

#1048 viirya closed 2 weeks ago
0
chore: Refactor Arrow Array and Schema allocation in ColumnReader and MetadataColumnReader

#1047 viirya closed 2 weeks ago
2
fix: need default value for getSizeAsMb(EXECUTOR_MEMORY.key)

#1046 neyama closed 3 weeks ago
1
Default value is required for sc.getConf.getSizeAsMb(EXECUTOR_MEMORY.key)

#1045 neyama closed 3 weeks ago
0
[EPIC] Add support for all Map functions

#1044 andygrove opened 3 weeks ago
0
[EPIC] Complex Type Support

#1043 andygrove opened 3 weeks ago
0
[EPIC] Add support for all array expressions

#1042 andygrove opened 3 weeks ago
0
chore: Use twox-hash 2.0 xxhash64 oneshot api instead of custom implementation

#1041 NoeB closed 3 weeks ago
11
Use parquet crate for decoding Parquet data into Arrow arrays

#1040 andygrove opened 3 weeks ago
5
feat: Support more types with BloomFilterAgg

#1039 mbutrovich closed 3 weeks ago
5
Some aggregate functions return 0.0 instead of NaN in some cases

#1038 andygrove opened 3 weeks ago
0
Cast from timestamp to decimal causes an exception

#1037 andygrove opened 3 weeks ago
1
cast negative zero to string inconsistent with Spark

#1036 andygrove opened 3 weeks ago
1
`CometBuffer` can potentially lead to concurrent modification of a held buffer (aka is "Unsound" in Rust terms)

#1035 tustvold opened 4 weeks ago
29
feat: Implement native version of ColumnarToRow

#1034 parthchandra closed 2 weeks ago
6
fix: TopK operator should return correct results on dictionary column with nulls

#1033 viirya closed 4 weeks ago
3
Use twox_hash 2.0

#1032 Dandandan closed 3 weeks ago
0
`TopK` operator (i.e. `CometTakeOrderedAndProjectExec`) may return incorrect result

#1030 viirya closed 4 weeks ago
2
perf: Cache jstrings during metrics collection

#1029 mbutrovich closed 3 weeks ago
8
Add support for Iceberg

#1028 andygrove opened 1 month ago
1
fix: Make comet-git-info.properties optional

#1027 andygrove closed 1 month ago
1
java.lang.ExceptionInInitializerError: Could not find comet-git-info.properties

#1026 BjarkeTornager closed 1 month ago
1
minor: Remove hard-coded version number from Dockerfile

#1025 andygrove closed 1 month ago
0
Reduce metrics collection overhead

#1024 mbutrovich opened 1 month ago
0
Add more types to BloomFilterAgg

#1023 mbutrovich closed 3 weeks ago
0
chore: Reserve memory for native shuffle writer per partition

#1022 viirya closed 1 month ago
5
feat: Add a `spark.comet.exec.memoryPool` configuration for experimenting with various datafusion memory pool setups.

#1021 Kontinuation opened 1 month ago
3
chore: Revert "chore: Reserve memory for native shuffle writer per partition (#988)"

#1020 viirya closed 1 month ago
1
CometNativeException: called `Option::unwrap()` on a `None` value

#1019 andygrove closed 1 month ago
2
Use unified memory management for Comet all the times

#1017 viirya closed 1 week ago
0
fix: Fallback to Spark if named_struct contains duplicate field names

#1016 viirya closed 1 month ago
1
Comet named_struct fails on duplicate field names

#1015 viirya closed 1 month ago
0
chore: remove legacy comet-spark-shell

#1013 andygrove closed 1 month ago
0
java.lang.NoClassDefFoundError: Could not initialize class org.apache.comet.package$

#1012 andygrove closed 1 month ago
2
[Research] Use custom cost model when deciding between SMJ and SHJ

#1011 andygrove opened 1 month ago
0
Include Apple OSX support in jars in Maven central

#1010 andygrove opened 1 month ago
3
docs: clarify that Maven central only has jars for Linux

#1009 andygrove closed 1 month ago
0
perf: Enable replaceSortMergeJoin by default

#1008 andygrove closed 1 month ago
0

Previous Next