Installation

Downloading the SimStack client

For Linux and Windows systems, the latest version of the SimStack client is available here. The client version installs itself and all dependencies during the first launch. The client version installs itself and all dependencies during the first launch. After extracting the .tar.gz or .zip files from the downloaded archive, run the proper commands according to your system.

For communicating with a compute backend, the client version requires passwordless via ssh access. If you do not have ssh access to the HPC resources already preconfigured, you need to generate a `ssh`keypair and transfer it to the Compute backend. You achieve this with two simple commands, as shown below.

  1. Installation on Linux:

    cd simstack_linux ./run_simstack.sh

Warning

If you don’t have the ssh keys, use the steps below to generate them..

  • SSH Key generation and copying to the HPC resource:

    ssh-keygen -t rsa
    
  • The ssh-key command in the step above generated two keys in the ~/.ssh directory

    id_rsa
    id_rsa.pub
    ssh-copy-id tutorialXX@int.bionano.int.kit.edu
    
  1. Installation on Windows:

    Double click on run-simstack.bat

    • This website has a clear explanation of how to generate a ssh key on Windows 10.

Warning

Make sure that the ssh is enabled on your Windows system.

Note

For this tutorial, the HPC resource considered here is the bionano cluster.

Simstack Server Configuration

  • Setup the server by opening the configuration menu: Configuration -> Servers

../_images/simstack_configuration.png
  • If ssh key is setup correctly, you should now be able to connect by clicking the Connect button at the top right of SimStack. The green button means you successfully connect to the server.

../_images/simstack_gui.png

Warning

If you get an error message during your try, double-check the field on the server setup.

Simstack Overview

../_images/simstack_overview.png

SimStack’s basic graphical user interface elements.

Using the SimStack client (picture above), simulation workflows are constructed by dragging and dropping various (already incorporated modules) from the window on the left side area (Available WaNos) into the Workflow canvas area. Double click each module to modify module-specific parameters (see input file field) and allocate resources in the Requested computational resources field for each module. To save and reuse your workflow lately, press ` Crtl+S` or `File -> Save`. It will then appear in the left panel Saved Workflows and can be re-loaded by double-clicking. To submit your workflow, connect to the computational resource (the connect button as shown in the last figure of Simstack Server Configuration section) and click `Run -> Run` on the menu bar, or `Crtl + r`. All required input files are uploaded automatically to the HPC resource, and workflow modules may run serially or in parallel, depending on if your workflow uses or not some of the Loop controls features. As shown in the figure above, the SimStack client will display a yellow folder while be running. When successfully finished, the client will exhibit a green folder, and you will be able to retrieve all the relevant data from your simulations. If the simulation presents a computational issue during the execution, SimStack returns a red folder, which we can be inspected to fix the problem.