This article describes how to delete data in Azure Cosmos DB Cassandra API tables from Spark.

Cassandra API configuration

import org.apache.spark.sql.cassandra._
//Spark connector
import com.datastax.spark.connector._
import com.datastax.spark.connector.cql.CassandraConnector
//CosmosDB library for multiple retry
import com.microsoft.azure.cosmosdb.cassandra
//Connection-related
spark.conf.set("spark.cassandra.connection.host","YOUR_ACCOUNT_NAME.cassandra.cosmosdb.azure.com")
spark.conf.set("spark.cassandra.connection.port","10350")
spark.conf.set("spark.cassandra.connection.ssl.enabled","true")
spark.conf.set("spark.cassandra.auth.username","YOUR_ACCOUNT_NAME")
spark.conf.set("spark.cassandra.auth.password","YOUR_ACCOUNT_KEY")
spark.conf.set("spark.cassandra.connection.factory", "com.microsoft.azure.cosmosdb.cassandra.CosmosDbConnectionFactory")
//Throughput-related...adjust as needed
spark.conf.set("spark.cassandra.output.batch.size.rows", "1")
spark.conf.set("spark.cassandra.connection.connections_per_executor_max", "10")
spark.conf.set("spark.cassandra.output.concurrent.writes", "1000")
spark.conf.set("spark.cassandra.concurrent.reads", "512")
spark.conf.set("spark.cassandra.output.batch.grouping.buffer.size", "1000")
spark.conf.set("spark.cassandra.connection.keep_alive_ms", "600000000")

Sample data generator

We will use this code fragment to generate sample data:

//Create dataframe
val booksDF = Seq(
   ("b00001", "Arthur Conan Doyle", "A study in scarlet", 1887,11.33),
   ("b00023", "Arthur Conan Doyle", "A sign of four", 1890,22.45),
   ("b01001", "Arthur Conan Doyle", "The adventures of Sherlock Holmes", 1892,19.83),
   ("b00501", "Arthur Conan Doyle", "The memoirs of Sherlock Holmes", 1893,14.22),
   ("b00300", "Arthur Conan Doyle", "The hounds of Baskerville", 1901,12.25)
).toDF("book_id", "book_author", "book_name", "book_pub_year","book_price")
//Persist
booksDF.write
  .mode("append")
  .format("org.apache.spark.sql.cassandra")
  .options(Map( "table" -> "books", "keyspace" -> "books_ks", "output.consistency.level" -> "ALL", "ttl" -> "10000000"))
  .save()

Dataframe API

Delete rows that match a condition

//1) Create dataframe
val deleteBooksDF = spark
  .read
  .format("org.apache.spark.sql.cassandra")
  .options(Map( "table" -> "books", "keyspace" -> "books_ks"))
  .load
  .filter("book_pub_year = 1887")
//2) Review execution plan
deleteBooksDF.explain
//3) Review table data before execution
println("==================")
println("1) Before")
deleteBooksDF.show
println("==================")
//4) Delete selected records in dataframe
println("==================")
println("2a) Starting delete")
//Reuse connection for each partition
val cdbConnector = CassandraConnector(sc)
deleteBooksDF.foreachPartition(partition => {
  cdbConnector.withSessionDo(session =>
    partition.foreach{ book => 
        val delete = s"DELETE FROM books_ks.books where book_id='"+book.getString(0) +"';"
        session.execute(delete)
    })
})
println("2b) Completed delete")
println("==================")
//5) Review table data after delete operation
println("3) After")
spark
  .read
  .format("org.apache.spark.sql.cassandra")
  .options(Map( "table" -> "books", "keyspace" -> "books_ks"))
  .load
  .show

Output:

== Physical Plan ==
*(1) Filter (isnotnull(book_pub_year#486) && (book_pub_year#486 = 1887))
+- *(1) Scan org.apache.spark.sql.cassandra.CassandraSourceRelation@197cfae4 [book_id#482,book_author#483,book_name#484,book_price#485,book_pub_year#486] 
PushedFilters: [IsNotNull(book_pub_year), EqualTo(book_pub_year,1887)], 
ReadSchema: struct<book_id:string,book_author:string,book_name:string,book_price:float,book_pub_year:int>
==================
1) Before
+-------+------------------+------------------+----------+-------------+
|book_id|       book_author|         book_name|book_price|book_pub_year|
+-------+------------------+------------------+----------+-------------+
| b00001|Arthur Conan Doyle|A study in scarlet|     11.33|         1887|
+-------+------------------+------------------+----------+-------------+
==================
==================
2a) Starting delete
2b) Completed delete
==================
3) After
+-------+------------------+--------------------+----------+-------------+
|book_id|       book_author|           book_name|book_price|book_pub_year|
+-------+------------------+--------------------+----------+-------------+
| b00300|Arthur Conan Doyle|The hounds of Bas...|     12.25|         1901|
| b03999|Arthur Conan Doyle|The adventure of ...|      null|         1892|
| b00023|Arthur Conan Doyle|      A sign of four|     22.45|         1890|
| b00501|Arthur Conan Doyle|The memoirs of Sh...|     14.22|         1893|
| b01001|Arthur Conan Doyle|The adventures of...|     19.83|         1892|
| b02999|Arthur Conan Doyle|  A case of identity|      15.0|         1891|
+-------+------------------+--------------------+----------+-------------+
deleteBooksDF: org.apache.spark.sql.Dataset[org.apache.spark.sql.Row] = [book_id: string, book_author: string ... 3 more fields]
cdbConnector: com.datastax.spark.connector.cql.CassandraConnector = com.datastax.spark.connector.cql.CassandraConnector@187deb43

Delete all the rows in the table

//1) Create dataframe
val deleteBooksDF = spark
  .read
  .format("org.apache.spark.sql.cassandra")
  .options(Map( "table" -> "books", "keyspace" -> "books_ks"))
  .load
//2) Review execution plan
deleteBooksDF.explain
//3) Review table data before execution
println("==================")
println("1) Before")
deleteBooksDF.show
println("==================")
//4) Delete records in dataframe
println("==================")
println("2a) Starting delete")
//Reuse connection for each partition
val cdbConnector = CassandraConnector(sc)
deleteBooksDF.foreachPartition(partition => {
  cdbConnector.withSessionDo(session =>
    partition.foreach{ book => 
        val delete = s"DELETE FROM books_ks.books where book_id='"+book.getString(0) +"';"
        session.execute(delete)
    })
})
println("2b) Completed delete")
println("==================")
//5) Review table data after delete operation
println("3) After")
spark
  .read
  .format("org.apache.spark.sql.cassandra")
  .options(Map( "table" -> "books", "keyspace" -> "books_ks"))
  .load
  .show

Output:

== Physical Plan ==
*(1) Scan org.apache.spark.sql.cassandra.CassandraSourceRelation@495377d7 [book_id#565,book_author#566,book_name#567,book_price#568,book_pub_year#569] 
PushedFilters: [], 
ReadSchema: struct<book_id:string,book_author:string,book_name:string,book_price:float,book_pub_year:int>
==================
1) Before
+-------+------------------+--------------------+----------+-------------+
|book_id|       book_author|           book_name|book_price|book_pub_year|
+-------+------------------+--------------------+----------+-------------+
| b00300|Arthur Conan Doyle|The hounds of Bas...|     12.25|         1901|
| b03999|Arthur Conan Doyle|The adventure of ...|      null|         1892|
| b00023|Arthur Conan Doyle|      A sign of four|     22.45|         1890|
| b00501|Arthur Conan Doyle|The memoirs of Sh...|     14.22|         1893|
| b01001|Arthur Conan Doyle|The adventures of...|     19.83|         1892|
| b02999|Arthur Conan Doyle|  A case of identity|      15.0|         1891|
+-------+------------------+--------------------+----------+-------------+
==================
==================
2a) Starting delete
2b) Completed delete
==================
3) After
+-------+-----------+---------+----------+-------------+
|book_id|book_author|book_name|book_price|book_pub_year|
+-------+-----------+---------+----------+-------------+
+-------+-----------+---------+----------+-------------+

RDD API

Delete all the rows in the table

//1) Create RDD with all rows
val deleteBooksRDD = 
    sc.cassandraTable("books_ks", "books")
//2) Review table data before execution
println("==================")
println("1) Before")
deleteBooksRDD.collect.foreach(println)
println("==================")
//3) Delete selected records in dataframe
println("==================")
println("2a) Starting delete")
/* Option 1: 
// Not supported currently
sc.cassandraTable("books_ks", "books")
  .where("book_pub_year = 1891")
  .deleteFromCassandra("books_ks", "books")
*/
//Option 2: CassandraConnector and CQL
//Reuse connection for each partition
val cdbConnector = CassandraConnector(sc)
deleteBooksRDD.foreachPartition(partition => {
    cdbConnector.withSessionDo(session =>
    partition.foreach{book => 
        val delete = s"DELETE FROM books_ks.books where book_id='"+ book.getString(0) +"';"
        session.execute(delete)
    }
   )
})
println("Completed delete")
println("==================")
println("2b) Completed delete")
println("==================")
//5) Review table data after delete operation
println("3) After")
sc.cassandraTable("books_ks", "books").collect.foreach(println)

Output:

==================
1) Before
CassandraRow{book_id: b00300, book_author: Arthur Conan Doyle, book_name: The hounds of Baskerville, book_price: 12.25, book_pub_year: 1901}
CassandraRow{book_id: b00001, book_author: Arthur Conan Doyle, book_name: A study in scarlet, book_price: 11.33, book_pub_year: 1887}
CassandraRow{book_id: b00023, book_author: Arthur Conan Doyle, book_name: A sign of four, book_price: 22.45, book_pub_year: 1890}
CassandraRow{book_id: b00501, book_author: Arthur Conan Doyle, book_name: The memoirs of Sherlock Holmes, book_price: 14.22, book_pub_year: 1893}
CassandraRow{book_id: b01001, book_author: Arthur Conan Doyle, book_name: The adventures of Sherlock Holmes, book_price: 19.83, book_pub_year: 1892}
==================
==================
2a) Starting delete
Completed delete
==================
2b) Completed delete
==================
3) After
deleteBooksRDD: com.datastax.spark.connector.rdd.CassandraTableScanRDD[com.datastax.spark.connector.CassandraRow] = CassandraTableScanRDD[126] at RDD at CassandraRDD.scala:19
cdbConnector: com.datastax.spark.connector.cql.CassandraConnector = com.datastax.spark.connector.cql.CassandraConnector@317927

Delete specific columns

//1) Create RDD 
val deleteBooksRDD = 
    sc.cassandraTable("books_ks", "books")
//2) Review table data before execution
println("==================")
println("1) Before")
deleteBooksRDD.collect.foreach(println)
println("==================")
//3) Delete specific column values
println("==================")
println("2a) Starting delete of book price")
sc.cassandraTable("books_ks", "books")
  .deleteFromCassandra("books_ks", "books",SomeColumns("book_price"))
println("Completed delete")
println("==================")
println("2b) Completed delete")
println("==================")
//5) Review table data after delete operation
println("3) After")
sc.cassandraTable("books_ks", "books").take(4).foreach(println)

Output:

==================
1) Before
CassandraRow{book_id: b00300, book_author: Arthur Conan Doyle, book_name: The hounds of Baskerville, book_price: 20.0, book_pub_year: 1901}
CassandraRow{book_id: b00001, book_author: Arthur Conan Doyle, book_name: A study in scarlet, book_price: 23.0, book_pub_year: 1887}
CassandraRow{book_id: b00023, book_author: Arthur Conan Doyle, book_name: A sign of four, book_price: 11.0, book_pub_year: 1890}
CassandraRow{book_id: b00501, book_author: Arthur Conan Doyle, book_name: The memoirs of Sherlock Holmes, book_price: 5.0, book_pub_year: 1893}
CassandraRow{book_id: b01001, book_author: Arthur Conan Doyle, book_name: The adventures of Sherlock Holmes, book_price: 10.0, book_pub_year: 1892}
==================
==================
2a) Starting delete of book price
Completed delete
==================
2b) Completed delete
==================
3) After
CassandraRow{book_id: b00300, book_author: Arthur Conan Doyle, book_name: The hounds of Baskerville, book_price: null, book_pub_year: 1901}
CassandraRow{book_id: b00001, book_author: Arthur Conan Doyle, book_name: A study in scarlet, book_price: null, book_pub_year: 1887}
CassandraRow{book_id: b00023, book_author: Arthur Conan Doyle, book_name: A sign of four, book_price: null, book_pub_year: 1890}
CassandraRow{book_id: b00501, book_author: Arthur Conan Doyle, book_name: The memoirs of Sherlock Holmes, book_price: null, book_pub_year: 1893}
deleteBooksRDD: com.datastax.spark.connector.rdd.CassandraTableScanRDD[com.datastax.spark.connector.CassandraRow] = CassandraTableScanRDD[145] at RDD at CassandraRDD.scala:19

Next steps

To perform aggregation and data copy operations, refer -