---
---
Installing Java applications (on Puhti)
💬 Java applications do not typically require special installation.
You just need to download the distribution package, typically a
.jar
file.
💡 Compute nodes do not have a default Java installation, so you need to load a suitable Java module.
You can check the available modules with the command:
💬 Despite their name, modules named biojava
are just normal Java installations, which can be used with any software, not just bio applications.
☝🏻 Some java applications fail to run on the login nodes. Try running them using sinteractive
instead (or by opening a compute node shell in the Puhti web interface).
‼️ Naturally, computationally heavy tasks should be never ran on the login nodes.
💡 The compute nodes have very limited /tmp
space. If you have to set the temporary file location for a Java application, you can use:
Example: Installing ABRA2
💬 In this example we'll install and run the Java application ABRA2.
Create a personal folder (if not done already) under your project's
/projappl
directory and move there. Then create (and move to) a new directory for the installation:Download the software:
Load a Java module:
💡 If you get error messages or warnings about the Java version, try loading another Java module.
Run the application:
💡 If you are not in the same directory as the
.jar
file, you need to include the full path to the file, e.g.
Example 2: Java application with a shell wrapper script
💬 Some Java applications come with shell scripts that are used to launch them instead of calling Java directly. See the documentation of each software for details.
💡 One example of such an application is BEAST:
☝🏻 BEAST will fail to run on the login nodes. Try running it in an interactive session instead (using sinteractive
or a compute node shell in the Puhti web interface).