Re: Spark is only using one worker machine when more are available

2018-04-12 Thread Gourav Sengupta
>> Dataset jdbcDF = ss.read().jdbc(this.url, dt, connProp); >> jdbcDF.createOrReplaceTempView(tableInfo.tmp_table_name); >> } >> } >> >> >> // Then run a query and write the result set to mysql >&g

Re: Spark is only using one worker machine when more are available

2018-04-12 Thread Jhon Anderson Cardenas Diaz
connProp.put("rewriteBatchedStatements", "true"); > connProp.put("sessionVariables", "sql_log_bin=off"); > result.write().jdbc(this.dst_url, this.dst_table, connProp); > > > >

Re: Spark is only using one worker machine when more are available

2018-04-11 Thread Jhon Anderson Cardenas Diaz
Hi, could you please share the environment variables values that you are sending when you run the jobs, spark version, etc.. more details. Btw, you should take a look on SPARK_WORKER_INSTANCES and SPARK_WORKER_CORES if you are using spark 2.0.0