Workshop Scope

Heterogeneity has become a defining feature of modern parallel and distributed computing systems, driven by the need for higher performance and better energy efficiency. It spans AI-focused supercomputers, classical HPC platforms, cloud and edge infrastructures, and increasingly complex memory and storage hierarchies. Effectively exploiting these diverse architectures requires coordinated advances in algorithms, programming models, runtime systems, compilers, and accurate performance and energy modeling. 

HeteroPar 2026 aims to provide a dedicated forum for researchers working on algorithms, programming languages, tools, and theoretical models aimed at efficiently solving problems on heterogeneous platforms. Topics to be covered include but are not limited to: 

• Heterogeneous parallel programming paradigms and models
• Languages, libraries, and interfaces for heterogeneous programming (e.g., SYCL, OpenMP/OpenACC offloading, task-based models)
• Performance and energy models; autotuning and adaptive runtimes for heterogeneous platforms - Parallel algorithms and scheduling for heterogeneous/hierarchical systems and accelerators (GPUs, FPGAs, AI accelerators)
• Parallel algorithms for efficient problem solving on heterogeneous platforms (e.g. numerical linear algebra, nonlinear systems, fast transforms, computational biology, data mining, artificial intelligence, multimedia)
• Applications and experience reports in HPC, data analytics, and AI on heterogeneous systems
• Algorithms, models and tools for energy and/or multi-objective optimization on heterogeneous platforms - Integration of parallel and distributed computing on heterogeneous systems
• Experience of porting parallel software from supercomputers to heterogeneous platforms - Resilience and fault tolerance for heterogeneous computing
• Algorithms, models and tools for grid, desktop grid, cloud, and green computing that include heterogeneous computing aspects 
• LLM- and AI-assisted programming, code generation, autotuning, and optimization for heterogeneous platforms

Authors are encouraged to submit original, unpublished research or overviews addressing the topics above.