Github user zzcclp commented on a diff in the pull request: https://github.com/apache/carbondata/pull/1455#discussion_r148963119 --- Diff: integration/spark2/src/main/scala/org/apache/spark/sql/execution/command/management/LoadTableCommand.scala --- @@ -84,6 +84,10 @@ case class LoadTableCommand( val carbonProperty: CarbonProperties = CarbonProperties.getInstance() carbonProperty.addProperty("zookeeper.enable.lock", "false") + carbonProperty.addProperty(CarbonCommonConstants.NUM_CORES_LOADING, + carbonProperty.getProperty(CarbonCommonConstants.NUM_CORES_LOADING, + Math.min(sparkSession.sparkContext.conf.getInt("spark.executor.cores", 1), + CarbonCommonConstants.NUM_CORES_MAX_VAL).toString())) --- End diff -- Can't modify *NUM_CORES_DEFAULT_VAL* to 32 directly, there are some places to use *NUM_CORES_DEFAULT_VAL*, for example: in org.apache.carbondata.core.datastore.BlockIndexStore.getAll: `try {` ` numberOfCores = Integer.parseInt(CarbonProperties.getInstance()` ` .getProperty(CarbonCommonConstants.NUM_CORES,` ` CarbonCommonConstants.NUM_CORES_DEFAULT_VAL));` ` } catch (NumberFormatException e) {` ` numberOfCores = Integer.parseInt(CarbonCommonConstants.NUM_CORES_DEFAULT_VAL);` ` }` 32 is too big.
---