home.social
  1. My implementation of the GaussSeidel smoother using a Diagnol direct access scheme in OpenFOAM as compared to the default GS smoother LDU gives almost ~50% improvements in cache misses and hits for a structured 3D cavity tutorial. Profiled using the amazing LIKWID profiler. Will share a deep technical report soon. Check it out and use - github.com/amartyadav/DIAGauss

  2. Released my DIA-format Gauss-Seidel smoother plugin for OpenFOAM v13. MIT licensed.

    Replaces the default LDU smoother on structured hex meshes — DIA stores diagonal bands contiguously, reducing pointer indirection and DRAM pressure. Expecting 10–20% wall-clock gains and better cache utilisation based on standalone profiling. Full OpenFOAM benchmarks incoming.

    github.com/amartyadav/DIAGauss

  3. Been working on my Linux machine for 3 months now at work - comp. physics, hpc, etc.
    Tried working on my MBP yesterday (after around 5 months of not using it for development) - while a superior machine h/w wise, I really couldn't stand macOS.
    Linux just works - better even.

  4. My latest blog post explores a Multi-Material Thermal Analysis using HYPRE BoomerAMG with a custom POSIT accelerator called Uttunga, and compares it to the x86 IEEE 754 standard.

    amartyadav.com/posts/post3.html

  5. My work for the month is to optimise the matrix adressing scheme of OpenFOAM to reduce cache miss. Initial idea is to replace the LDU Matrix addressing scheme with Diagnol matrix addressing scheme using multiple arrays for structures meshes. Will check it out on a simple 2D Poisson equation using Gauss-Seidel solver to check performance benefits, and will eventually proceed to introduce a new solver as a plugin if benefits are noticeable.
    Wish me luck.

  6. Might publish a blog or two in a few weeks about my work with OpenFOAM, different profilers, and an implementation of an LBM CFD kernel on RISC-V with POSITS.

    What a great company to be working for. Learning so much!

  7. This is me after with some of the attendees after successfully delivering my first workshop at the India 2025 in Bengaluru.
    Taught how to perform an "Efficient Halo Exchange for stencil-based codes on the Wafer-Scale Engine 3"

  8. After four months of deep work in computational physics and wafer-scale computing using the Cerebras WSE-3 as part of my MSc dissertation, my tutorial has been accepted at Supercomputing India 2025!” in Bengaluru.
    Excited to contribute and share practical insights with the HPC community.

  9. The struggle and perseverance to get this output from the Cerebras CSL compiler 😌

  10. Looking to connect with researchers/programmers using the Cerebras WSE for scientific programming. I’m currently implementing CFD on Cerebras WSE-3 as part of my MSc project.

  11. My brain is absolutely fried.
    Today is the last day of coursework submissions for this semester. What a hectic month.
    DNN with PyTorch, Brain model parallelisation with MPI, SYCL and OpenMP offloading of percolation models,hand optimizing serial codes for performance.
    Two submissions due today. Submitted one and finalising my report for the second one.
    Definitely having a pint after this

  12. Started SYCL this semester in my MSc, and I have a coursework on it.
    I have never been more frustrated in my life.
    I am not saying SYCL is bad. I might just be too dumb to master it in a sem in order to port an existing CPU code to use MPI & SYCL together.
    CUDA was much easier for me for the same task.

  13. I’ve started listening to traditional radio again.
    This is primarily because of the stress and anxiety that news from social media and news portals gives you.
    Listening to 15 minutes of morning, afternoon and evening news on Indian national radio (Aakashvaani) keeps me updated and stress-free.

  14. Listening to “Lekin” by JJ47 while running experiments using OpenMP on Cirrus as part of my coursework. Nice Saturday evening.

  15. Made it to the team - TeamEPCC for the Student Cluster Competiton at the International Supercomputing Conference 2025 in Hamburg, Germany!

  16. Submitted my application for TeamEPCC for the Student Cluster Competition at ISC 2025, Hamburg, Germany. Fingers crossed.

  17. Starting to prepare my application for TeamEPCC for the Student Cluster Competition at ISC 2025, Hamburg, Germany.

    Let's see if I can make it to the team haha! 🙏