Accessing OpenStack Swift from Spark

Spark’s support for Hadoop InputFormat allows it to process data in OpenStack Swift using the same URI formats as in Hadoop. You can specify a path in Swift as input through a URI of the form swift://container.PROVIDER/path. You will also need to set your Swift security credentials, through core-site.xml or via SparkContext.hadoopConfiguration. Current Swift driver requires Swift to use Keystone authentication method.

Configuring Swift for Better Data Locality

Although not mandatory, it is recommended to configure the proxy server of Swift with list_endpoints to have better data locality. More information is available here.

Dependencies

The Spark application should include hadoop-openstack dependency. For example, for Maven support, add the following to the pom.xml file:

<dependencyManagement>
  ...
  <dependency>
    <groupId>org.apache.hadoop</groupId>
    <artifactId>hadoop-openstack</artifactId>
    <version>2.3.0</version>
  </dependency>
  ...
</dependencyManagement>

Configuration Parameters

Create core-site.xml and place it inside Spark’s conf directory. There are two main categories of parameters that should to be configured: declaration of the Swift driver and the parameters that are required by Keystone.

Configuration of Hadoop to use Swift File system achieved via

Property Name	Value
fs.swift.impl	org.apache.hadoop.fs.swift.snative.SwiftNativeFileSystem

Additional parameters required by Keystone (v2.0) and should be provided to the Swift driver. Those parameters will be used to perform authentication in Keystone to access Swift. The following table contains a list of Keystone mandatory parameters. PROVIDER can be any name.

Property Name	Meaning	Required
`fs.swift.service.PROVIDER.auth.url`	Keystone Authentication URL	Mandatory
`fs.swift.service.PROVIDER.auth.endpoint.prefix`	Keystone endpoints prefix	Optional
`fs.swift.service.PROVIDER.tenant`	Tenant	Mandatory
`fs.swift.service.PROVIDER.username`	Username	Mandatory
`fs.swift.service.PROVIDER.password`	Password	Mandatory
`fs.swift.service.PROVIDER.http.port`	HTTP port	Mandatory
`fs.swift.service.PROVIDER.region`	Keystone region	Mandatory
`fs.swift.service.PROVIDER.public`	Indicates if all URLs are public	Mandatory

For example, assume PROVIDER=SparkTest and Keystone contains user tester with password testing defined for tenant test. Then core-site.xml should include:

<configuration>
  <property>
    <name>fs.swift.impl</name>
    <value>org.apache.hadoop.fs.swift.snative.SwiftNativeFileSystem</value>
  </property>
  <property>
    <name>fs.swift.service.SparkTest.auth.url</name>
    <value>http://127.0.0.1:5000/v2.0/tokens</value>
  </property>
  <property>
    <name>fs.swift.service.SparkTest.auth.endpoint.prefix</name>
    <value>endpoints</value>
  </property>
    <name>fs.swift.service.SparkTest.http.port</name>
    <value>8080</value>
  </property>
  <property>
    <name>fs.swift.service.SparkTest.region</name>
    <value>RegionOne</value>
  </property>
  <property>
    <name>fs.swift.service.SparkTest.public</name>
    <value>true</value>
  </property>
  <property>
    <name>fs.swift.service.SparkTest.tenant</name>
    <value>test</value>
  </property>
  <property>
    <name>fs.swift.service.SparkTest.username</name>
    <value>tester</value>
  </property>
  <property>
    <name>fs.swift.service.SparkTest.password</name>
    <value>testing</value>
  </property>
</configuration>

Notice that fs.swift.service.PROVIDER.tenant, fs.swift.service.PROVIDER.username, fs.swift.service.PROVIDER.password contains sensitive information and keeping them in core-site.xml is not always a good approach. We suggest to keep those parameters in core-site.xml for testing purposes when running Spark via spark-shell. For job submissions they should be provided via sparkContext.hadoopConfiguration.