Emerging data-intensive, supercomputing applications are moving towards a convergence of scientific simulations, data analytics, and learning algorithms. Many of the components of these applications pertain both to well established and emerging fields, such as machine learning, social network analysis, bioinformatics, semantic graph databases, Computer Aided Design (CAD), and computer security. In processing massive sets of unstructured data, the components often execute many irregular, fine-grain accesses and synchronization events. Since current high-performance programming models, runtimes, and architectures rely on regular task graphs, bulk synchronous communications and high temporal and spatial data locality to reduce operational latencies, it is difficult to express irregular applications in current HPC programming models and scale performance on current supercomputing machines. Development of improved programming and execution models that address the issues of irregular applications is critical to solving the data challenges in large-scale science and data analysis.

This workshop seeks to explore solutions for supporting efficient execution of irregular applications in the form of new features at the level of the micro- and system architecture, network, languages and libraries, runtimes, compilers, algorithms, and performance studies.

Special subtopic: new for this year, the workshop will also host a special subtopic focused on dynamic graphs. This subtopic invites experts from various disciplines to share insights and advancements in the field of dynamic network analysis, looking to address  innovative methods, applications, and challenges related to large dynamic graphs.

Topics of interest, of both theoretical and practical significance, include but are not limited to:

 

- Micro- and System-architectures, including multi- and many-core designs, heterogeneous processors, accelerators (GPUs, vector processors, Automata processor, AI/ML accelerators), reconfigurable (coarse grained reconfigurable and FPGA designs) and custom processors 

- Network architectures and interconnects including high-radix and optical networks 

- Novel memory architectures and designs (including processors-in memory) 

- Impact of new computing paradigms on irregular workloads (including neuromorphic processors and quantum computing) 

- Modeling, simulation, and evaluation of novel architectures with irregular workloads 

- Languages and programming models for irregular workloads

- Library and runtime support for irregular workloads

- Compiler and analysis techniques for irregular workloads

- Innovative algorithmic techniques for irregular workloads

- Combinatorial algorithms (graph algorithms, sparse linear algebra, etc.)

- Impact of irregularity on machine learning approaches (e.g., graph neural networks, large language models)

- Parallelization techniques and data structures for irregular workloads

- Data structures combining regular and irregular computations (e.g., attributed graphs)

- Approaches for managing massive unstructured datasets (including streaming data) 

- High performance data analytics applications (including graph databases and solutions that combine graph algorithms with machine learning)

- Applications that integrate scientific simulation, data analytics, and learning, and require efficient execution of irregular workloads

- Hardware and software platforms, parallel algorithms, benchmarking, applications for dynamic graphs and dynamic graph neural networks

 

The workshop welcomes regular paper submissions, papers describing work-in-progress or incomplete but sound, as well as innovative ideas related to the workshop theme. We solicit both 8-page regular papers and 4-page position papers. Authors of exciting but not mature enough regular papers may be offered the option of a short 4-page paper and related short presentation.