Basic Concepts and Terminology
In the context of Cloudera Machine Learning, engines are responsible for running data science workloads and intermediating access to the underlying cluster. This topic walks you through some basic concepts and terminology related to engines.
Cloudera Machine Learning uses Docker containers to deliver application components and run isolated user workloads. On a per project basis, users can run R, Python, and Scala workloads with different versions of libraries and system packages. CPU and memory are also isolated, ensuring reliable, scalable execution in a multi-tenant setting.
Cloudera Machine Learning engines are responsible for running R, Python, and Scala code written by users. You can think of an engine as a virtual machine, customized to have all the necessary dependencies while keeping each project’s environment entirely isolated.
To enable multiple users and concurrent access, Cloudera Machine Learning transparently subdivides and schedules containers across multiple hosts. This scheduling is done using Kubernetes, a container orchestration system used internally by Cloudera Machine Learning. Neither Docker nor Kubernetes are directly exposed to end users, with users interacting with Cloudera Machine Learning through a web application.
- Base Engine Image
The base engine image is a Docker image that contains all the building blocks needed to launch a Cloudera Machine Learning session and run a workload. It consists of kernels for Python, R, and Scala along with additional libraries that can be used to run common data analytics operations. When you launch a session to run a project, an engine is kicked off from a container of this image. The base image itself is built and shipped along with Cloudera Machine Learning.
Cloudera Machine Learning offers legacy engines and Machine Learning Runtimes. Both legacy engines and ML Runtimes are Docker images and contain OS, interpreters, and libraries to run user code in sessions, jobs, experiments, models, and applications. However, there are significant differences between these choices. See ML Runtimes versus Legacy Engines for a summary of these differences.
New versions of the base engine image are released periodically. However, existing projects are not automatically upgraded to use new engine images. Older images are retained to ensure you are able to test code compatibility with the new engine before upgrading to it manually.
The term engine refers to a virtual machine-style environment that is created when you run a project (via session or job) in Cloudera Machine Learning. You can use an engine to run R, Python, and Scala workloads on data stored in the underlying CDH cluster.
Cloudera Machine Learning allows you to run code using either a session or a job. A session is a way to interactively launch an engine and execute code while a job lets you batch process those actions and schedule them to run recursively. Each session and job launches its own engine that lives as long as the workload is running (or until it times out).A running engine includes the following components:
Each engine runs a kernel with an R, Python or Scala process that can be used to execute code within the engine. The kernel launched differs based on the option you select (either Python 2/3, PySpark, R, or Scala) when you launch the session or configure a job.
The Python kernel is based on the Jupyter IPython kernel; the R kernel is custom-made for CML; and the Scala kernel is based on the Apache Toree kernel.
Project Filesystem Mount
Cloudera Machine Learning uses a persistent filesystem to store project files such as user code, installed libraries, or even small data files. Project files are stored on the master host at
Every time you launch a new session or run a job for a project, a new engine is created ,and the project filesystem is mounted into the engine's environment at
/home/cdsw. Once the session/job ends, the only project artifacts that remain are a log of the workload you ran, and any files that were generated or modified, including libraries you might have installed. All of the installed dependencies persist through the lifetime of the project. The next time you launch a session/job for the same project, those dependencies will be mounted into the engine environment along with the rest of the project filesystem.
If there are any files on the hosts that should be mounted into the engines at launch time, use the Site Administration panel to include them.
For detailed instructions, see Configuring the Engine Environment.