clear query| facets| time Search criteria: author:"Wenchen Fan".   Results from 31 to 40 from 649 (0.0s).
Loading phrases to help you
refine your search...
[SPARK-10743] keep the name of expression if possible when do cast - Spark - [issue]    Author: Wenchen Fan , 2015-10-21, 20:30
[SPARK-10765] use new aggregate interface for hive UDAF - Spark - [issue]    Author: Wenchen Fan , 2016-03-10, 22:52
[SPARK-10585] only copy data once when generate unsafe projection - Spark - [issue]
...When we have nested struct, array or map, we will create a byte buffer for each of them, and copy data to the buffer first, then copy them to the final row buffer. We can save the first copy...    Author: Wenchen Fan , 2015-10-06, 17:04
[SPARK-6138] Enhance the `toArray` method in `SizeTrackingVector` - Spark - [issue]
...Use array copy instead of `Iterator#toArray` to make it more efficient....    Author: Wenchen Fan , 2015-03-03, 12:12
[SPARK-6394] cleanup BlockManager companion object and improve the getCacheLocs method in DAGScheduler - Spark - [issue]
...The current implementation of getCacheLocs include searching a HashMap many times, we can avoid this.And BlockManager.blockIdsToExecutorIds isn't called anywhere, we can remove it. Also we c...    Author: Wenchen Fan , 2015-03-19, 02:43
[SPARK-1888] enhance MEMORY_AND_DISK mode by dropping blocks in parallel - Spark - [issue]
...Sometimes MEMORY_AND_DISK mode is slower than DISK_ONLY mode because of the lock on IO operations(dropping blocks in memory store). As the TODO says, the solution is: only synchronize the se...    Author: Wenchen Fan , 2014-12-11, 20:39
[SPARK-1912] Compression memory issue during reduce - Spark - [issue]
...When we need to read a compressed block, we will first create a compress stream instance(LZF or Snappy) and use it to wrap that block.Let's say a reducer task need to read 1000 local shuffle...    Author: Wenchen Fan , 2014-08-28, 08:07
[SPARK-10389] support order by non-attribute grouping expression on Aggregate - Spark - [issue]
...For example, we should support "SELECT MAX(value) FROM src GROUP BY key + 1 ORDER BY key + 1"....    Author: Wenchen Fan , 2015-10-13, 23:17
[SPARK-10154] remove the no-longer-necessary CatalystScan - Spark - [issue]    Author: Wenchen Fan , 2016-10-08, 05:51
[SPARK-10174] refactor out project, filter, ordering generator from SparkPlan - Spark - [issue]    Author: Wenchen Fan , 2016-10-08, 05:55