Sciunit
Sciunit is a command-line tool which creates efficient, lightweight, self-contained packages of computational experiments that can be guaranteed to reproduce. Sciunit creates a reusable research object that containerizes and stores applications, facilitates sharing and collaboration, and eases the task of executing, understanding, and building on shared work.
Sciunit Workflow
A Sciunit container consists of multiple executions. Each execution refers to running a command under Linux. The command may be a single binary, may start with the name of a specific virtual machine for managed languages such as “java,” or may be a shell script that contains multiple commands.
Sciunit uses application virtualization to capture and repeat an execution and works in two modes: audit and repeat. During audit mode, the dependencies of an execution are determined including its code, data, and environment, which are saved to a container. During repeat phase, the containerized execution is repeated on the same or different machine at a later stage.
Core Sciunit Commands
sciunit create <name>
Create a new Sciunit project under ~/sciunit/<name> and open it.
sciunit open <name>|<url>
Open the Sciunit project under ~/sciunit/<name> or designated by a url.
sciunit exec <executable> <args...>
Capture the execution of the given executable with the command line arguments args. The newly-created execution is added to the currently-opened Sciunit project and assigned execution id eN, where N is a monotonically-increasing decimal.
sciunit repeat <execution id>
Repeat the execution of execution id from the currently-opened Sciunit project exactly as it happened earlier.
sciunit list
List the existing executions in the currently-opened Sciunit project.
sciunit show <execution id>
Show detailed information about a specific execution in the currently-opened Sciunit project.
FLINC
FLINC is built on top of Sciunit. It enables users to audit and repeat programs executing in interactive notebook environments like Jupyter. FLINC creates two new kernels: 1. Audit Kernel: executes, audits, and containerizes notebook code 2. Repeat Kernel: repeats the containerized code
Hands-on progression
1. Sciunit Quickstart
Start with the Sciunit Quickstart. You will use Sciunit to run basic Linux commands like pwd and date, and learn how to run a very simple hello world script in bash.
2. Executing a Workflow with Sciunit
Move on to Workflow Execution with Sciunit. In this example, you will run a multi-step workflow using Sciunit audit and then reproduce it using Sciunit repeat.
3. Executing an Interactive Workflow with FLINC
Finally, you will open Notebook Execution with FLINC. In this example, you will execute the same workflow as #2 above in an interactive Jupyter notebook environment.
These exercises demonstrate how to create a reproducible execution in a shell environment using Sciunit and in an interactive environment using FLINC.