Our Services

Hands-on training in performance tuning and methodology transfer.

Performance Engineering

Master best practices for profiling and tuning applications.

Auto-Tuning Techniques

Learn automated methods to optimize your workflows efficiently.

Methodology Transfer

Equip your team with cutting-edge optimization strategies.

IT Services Portfolio

Hands-on advanced tuning workshops

Intensive 2–5 day training sessions covering modern performance engineering techniques: roofline analysis, cache-aware programming, vectorization deep-dive, NUMA optimization, GPU kernel tuning, and multi-backend portability (OpenMP target, SYCL, HIP)

Customized in-house methodology transfer

Roofline & performance portability masterclasses

Parallel programming certification pathway

Profiling & tooling mastery bootcamps

Long-term coaching & knowledge retention

Tailored knowledge transfer programs where DML HPC engineers work side-by-side with your team on real production codes to embed HPC optimization best practices directly into your internal workflows

Training on applying the Roofline model, computational intensity optimization, algorithmic redesign, and performance-portable frameworks (Kokkos, RAJA, oneAPI) to achieve consistent high performance across CPU, GPU, and accelerator architectures

Deep training on state-of-the-art tools: Nsight Systems/Compute, rocprof, LIKWID, TAU, Score-P, Vampir, Arm Forge, Intel VTune, gprof-ng — with hands-on labs to interpret traces, identify true bottlenecks, and eliminate false positives

Structured multi-level curriculum (beginner → advanced → expert) leading to internal certification in MPI, OpenMP 5+, CUDA/HIP, SYCL, and heterogeneous programming, with DML HPC-issued completion certificates

Post-training coaching program (quarterly deep-dive sessions, code review support, on-demand Q&A) to ensure your team maintains and evolves optimization skills, including updates on new hardware (e.g., NVIDIA Blackwell, AMD MI400, Intel Panther Lake) and emerging standards