Although we have covered quite a lot in this chapter, there is still a lot to be covered with Data Pipeline. One of the fastest and easiest ways to get started with Data Pipeline is by using one of the ready made pipeline definition templates.

As of date, Data Pipeline provides the following list of ready-to-use templates, using which you can get started with your own pipeline in a matter of minutes:

You can additionally use these definitions as templates for further customizing and enhancing your own as well. Simply create a pipeline using one of the previously depicted templates, however do not activate them. Edit the pipeline in the architect mode and simple export the pipeline definition locally. Once the template's pipeline definition is saved locally, you can make further changes and enhancements to it or simply reuse components within it to make your own pipeline, as well. The possibilities are endless!

Another cool feature provided by pipelines is the use of spot instances as task nodes. By default, pipelines provide only on-demand instances as resources for your task nodes. You can optionally switch to spot instances by simply selecting the Task instance Bid Price option from the Resources pane of your pipeline. Provide a suitable amount in the adjoining field (between 0 and 20.00) and there you have it! The next time the pipeline activates and a task is run, it will be performed based on the availability of spot instances.