Debugging HPC Applications with mdb

The problem:

The current landscape of Free and Open-Source (FOS) MPI-debuggers is bleak. There are proprietary options available (e.g., linaro ddt, TotalView), but these require paid-for licenses. As a research software engineer, I work with my collaborators on whichever systems they have access to -- meaning that I frequently move from one HPC system to another. Some HPC service providers simply cannot afford the license costs and may provide one tool, the other or none. As a result relying on these tools can leave you in the dark when they are no longer available. Furthermore, each of these proprietary tools has a different interface and commands that would need to be re-learnt when you move to a different system.

My proposed solution:

mdb is a new FOS debugger for MPI applications, supporting Rust, C, C++ and Fortran. It is essentially a wrapper around CLI debuggers e.g., gdb, lldb etc. The GNU debugger gdb is a de facto standard for serial program debugging. Due to the ubiquity of gdb, mdb allows users to re-use their existing knowledge of debugging commands to debug MPI code on pretty much any system they want to run on. mdb is mostly a superset of gdb providing most of the core-functionality with some parallel-specific additions that help when debugging MPI applications.

In this talk, I will explain why I built mdb, how it works and I will provide a live-demo of how to debug a simple MPI code written in rust.

New AMD support:

I am excited to share that I recently added support for rocgdb which allows users to debug MPI-aware kernels and memory on AMD GPUs e.g., MI300's.

mdb Github repository