Hive Incremental on Partitioned table

Question

I am working on implementing the incremental process on hive table A; Table A - is already created in hive with partitioned on YearMonth ( YYYYMM column ) with full volume.

On-going basis, we are planning to import the updates/inserts from source and capture in hive Delta Table;

as shown in below picture, Delta table indicates that new updates are pertaining to partitions ( 201804 / 201611 / 201705 ).

For incremental process , I am planning to

1. Select 3 Partitions from original table which are affected.

INSERT INTO delta2 select YYYYMM from Table where YYYYMM in ( select distinct YYYYMM from Delta );

2. Merge these 3 partitions from Delta table with corresponding partitions from original table. ( I can follow Horton works 4 step strategy to apply the updates )

       Merge Delta2 + Delta : = new 3 partitions.
   

3. Drop 3 partitions from original table

   Alter Table Drop partitions 201804 / 201611 / 201705
   

4. Add newly merged partitions back to Original table ( having new updates )

I need to automate this scripts - Can you please suggest how to put above logic in hive QL or spark - Speacifically Identify partitions and drop them from original table.

Answer 1

you can build a solution using pyspark. I am explaining this approach with some basic example. you can re-modify it as per your business requirements.

Suppose you have a partitioned table in hive below configuration.

CREATE TABLE IF NOT EXISTS udb.emp_partition_Load_tbl (
 emp_id                 smallint
,emp_name               VARCHAR(30)
,emp_city               VARCHAR(10)
,emp_dept               VARCHAR(30)
,emp_salary             BIGINT
)
PARTITIONED BY (Year String, Month String)
ROW FORMAT DELIMITED FIELDS TERMINATED BY '|'
STORED AS ORC;

and you got some csv file with some input records which you want to load into your partitioned table

1|vikrant singh rana|Gurgaon|Information Technology|20000

dataframe = spark.read.format("com.databricks.spark.csv") \
  .option("mode", "DROPMALFORMED") \
  .option("header", "false") \
  .option("inferschema", "true") \
  .schema(userschema) \
  .option("delimiter", "|").load("file:///filelocation/userinput")

newdf = dataframe.withColumn('year', lit('2018')).withColumn('month',lit('01'))

+------+------------------+--------+----------------------+----------+----+-----+
|emp-id|emp-name          |emp-city|emp-department        |emp-salary|year|month|
+------+------------------+--------+----------------------+----------+----+-----+
|1     |vikrant singh rana|Gurgaon |Information Technology|20000     |2018|01   |
+------+------------------+--------+----------------------+----------+----+-----+

setting below properties to overwrite specific partitions data only.

spark.conf.set("spark.sql.sources.partitionOverwriteMode","dynamic")
spark.sql("set spark.hadoop.hive.exec.dynamic.partition=true");
spark.sql("set spark.hadoop.hive.exec.dynamic.partition.mode=nonstrict");

newdf.write.format('orc').mode("overwrite").insertInto('udb.emp_partition_Load_tbl')

lets say you got another set of data and want to insert into some other partitions

+------+--------+--------+--------------+----------+----+-----+
|emp-id|emp-name|emp-city|emp-department|emp-salary|year|month|
+------+--------+--------+--------------+----------+----+-----+
|     2|     ABC| Gurgaon|HUMAN RESOURCE|     10000|2018|   02|
+------+--------+--------+--------------+----------+----+-----+
newdf.write.format('orc').mode("overwrite").insertInto('udb.emp_partition_Load_tbl')

> show partitions udb.emp_partition_Load_tbl;
+---------------------+--+
|      partition      |
+---------------------+--+
| year=2018/month=01  |
| year=2018/month=02  |
+---------------------+--+

assuming you have got another set of records pertaining to existing partition.

3|XYZ|Gurgaon|HUMAN RESOURCE|80000

newdf = dataframe.withColumn('year', lit('2018')).withColumn('month',lit('02'))
+------+--------+--------+--------------+----------+----+-----+
|emp-id|emp-name|emp-city|emp-department|emp-salary|year|month|
+------+--------+--------+--------------+----------+----+-----+
|     3|     XYZ| Gurgaon|HUMAN RESOURCE|     80000|2018|   02|
+------+--------+--------+--------------+----------+----+-----+

newdf.write.format('orc').mode("overwrite").insertInto('udb.emp_partition_Load_tbl')


 select * from udb.emp_partition_Load_tbl where year ='2018' and month ='02';
+---------+-----------+-----------+-----------------+-------------+-------+--------+--+
| emp_id  | emp_name  | emp_city  |    emp_dept     | emp_salary  | year  | month  |
+---------+-----------+-----------+-----------------+-------------+-------+--------+--+
| 3       | XYZ       | Gurgaon   | HUMAN RESOURCE  | 80000       | 2018  | 02     |
| 2       | ABC       | Gurgaon   | HUMAN RESOURCE  | 10000       | 2018  | 02     |
+---------+-----------+-----------+-----------------+-------------+-------+--------+--+

you can see below that other partiion data was untouched.

> select * from udb.emp_partition_Load_tbl where year ='2018' and month ='01';

+---------+---------------------+-----------+-------------------------+-------------+-------+--------+--+
| emp_id  |      emp_name       | emp_city  |        emp_dept         | emp_salary  | year  | month  |
+---------+---------------------+-----------+-------------------------+-------------+-------+--------+--+
| 1       | vikrant singh rana  | Gurgaon   | Information Technology  | 20000       | 2018  | 01     |
+---------+---------------------+-----------+-------------------------+-------------+-------+--------+--+