Data Refinery environments

In Data Refinery, a Spark R runtime is started when:

  • You shape your data in Data Refinery
  • You run a Data Refinery flow in a job

All runtimes consume capacity unit hours (CUHs) that are tracked.

Shaping data in Data Refinery

When you select to refine data in Data Refinery, a Default Data Refinery XS runtime is started automatically and is listed as an active runtime on the Environments page of your project.

Remember: When you close the refinery interface, the runtime isn’t stopped immediately. You must stop the runtime from the Environments page of your project. If you don’t explicitly stop the runtime, it is stopped for you after an idle time of 1 hour.

Running a Data Refinery flow

You can create a job in which to run your Data Refinery flow:

  • Directly in Data Refinery by clicking the jobs icon from the Data Refinery toolbar and creating a job
  • From your project’s Jobs page
  • From your project’s Assets page by selecting the Data Refinery flow and clicking ACTIONS > Create job

Environment options in jobs

When you create a job in which to run a Data Refinery flow, you can select to use one the following environments:

  • A Spark R environment

    With a Spark R environment, the Data Refinery flow runs in its own Spark cluster. Each Spark environment consists of one SparkR kernel as a service. The kernel has a dedicated Spark cluster and Spark executors.

    You can select a Spark with R environment definition provided by Watson Studio or create your own Spark with R environment definition. If you create your own Spark with R environment definition, you can configure the size of the Spark driver and the size and number of the executors dependent on the size of the data set.

    You should always select a Spark with R environment to run Data Refinery flows that operate on large data sets.

    All Spark R environments are HIPAA ready.

  • The Default Data Refinery XS environment

    The Default Data Refinery XS runtime is used when you refine data in Data Refinery and can also be selected as the environment runtime when you create a job in which to run your Data Refinery flow.

    You should select the Default Data Refinery XS runtime to run Data Refinery flows that operate on small data sets because the runtime is instantly available and doesn’t first have to be started before the job can run.

    The Default Data Refinery XS runtime is HIPAA ready.

After the runtime was started, it is listed as an active runtime on the Environments page of your project. The runtime is stopped when the Data Refinery job stops running.

Default Spark with R environment definitions

Watson Studio offers the following Spark R environment definitions which you can select when you create a job in which to run a Data Refinery flow. Selecting this Spark environment definition helps you to quickly get started running Data Refinery jobs without having to create your own Spark R environment definition. The default environment definitions are listed on the project’s Environments page.

Name Hardware configuration
Default Spark 2.4 & R 3.6 2 Executors each: 1 vCPU and 4 GB RAM;
Driver: 1 vCPU and 4 GB RAM
Default Spark 2.3 & R 3.4 2 Executors each: 1 vCPU and 4 GB RAM;
Driver: 1 vCPU and 4 GB RAM

 

Runtime logs for jobs

To view the accumulated logs for a Data Refinery job:

  1. From the project’s Jobs page, click the job that ran the Data Refinery flow for which you want to see logs.
  2. Click the job run. You can view the log tail or download the complete log file.

FAQs

When should I use a Spark with R environment?

You should use a Spark with R environment only if you are working on a large data set. If your data set is small, you should select the Default Data Refinery XS runtime. The reason is that, although the SparkR cluster in a Spark R environment is fast and powerful, it requires time to create, which is noticeable when you run a Data Refinery job on small data set.

What is the difference between a Spark with R environment used in Data Refinery and the other Spark with R environments in a project?

These environments are the same.

When you create a job in which to run a Data Refinery flow, you can select a default Spark with R environment (either Default Spark 2.3 & R 3.4 or Default Spark 2.4 & R 3.6) or any other Spark R environment definition that you created.

Next steps