我有一个javapairdd，我想对它进行迭代，执行一些操作，并将输出存储到hive。目前，我正在尝试在foreach中创建一个dataframe，它抛出了一个异常，因为在foreach中不能创建dataframe。那么，还有什么选择呢？

JavaPairRDD<Long, Iterable<EmployeeDetail>>  employeeDetailPairList = fetchEmployeeDetailData();
        List<EmployeeZone> employeeZoneFCList = fetchEmployeeZoneData();
        employeeDetailPairList.foreach(employeeDetailPair ->  {
            Iterable<EmployeeDetail> employeeDetailList = employeeDetailPair._2;
            Set<String> zipCodeSet = StreamSupport.stream(employeeDetailList.spliterator(), false).map(e -> e.getZipCode()).collect(Collectors.toSet());
            List<EmployeeZone> employeeZoneFilteredList = employeeZoneList.stream().filter(e -> zipCodeSet.contains(String.valueOf(e.getLoc()))).collect(Collectors.toList());
            List<Output> outputListList = processEmployeeData(employeeZoneFilteredList);
            outputListList = addWeekStartDay(outputListList, weekStartDay);
            if(outputListList != null && this.getSession()!= null) {
              Dataset<Row> recordsDF = this.getSession().sqlContext().createDataFrame(outputListList, Output.class);
              recordsDF.write().insertInto(SHIPCODE_PREFERRED_FC_HIVE_TABLE);
            }
        });