Technical Paper Topic Areas 

Submissions will be considered on any topic related to high-performance computing including, but not limited to, the nine topical areas below. 

Algorithms: The development, evaluation and optimization of scalable, general-purpose, high-performance algorithms. 

Topics include: 

• Algorithmic techniques to improve energy and power efficiency 

• Algorithmic techniques to improve load balance 

• Data-intensive parallel algorithms 

• Discrete and combinatorial problems 

• Fault-tolerant algorithms 

• Graph algorithms 

• Hybrid/heterogeneous/accelerated algorithms 

• Network algorithms 

• Numerical methods, linear and nonlinear systems 

• Scheduling algorithms 

• Uncertainty quantification 

• Other high-performance algorithms 

Applications: The development and enhancement of algorithms, models, software and problem solving environments for domain-specific applications that require high-performance resources. 

Topics include: 

• Bioinformatics and computational biology 

• Computational earth and atmospheric sciences 

• Computational materials science and engineering 

• Computational astrophysics/astronomy, chemistry, and physics 

• Computational fluid dynamics and mechanics 

• Computation and data enabled social science 

• Computational design optimization for aerospace, energy, manufacturing and industrial applications 

• Computational medicine and bioengineering 

• Use of uncertainty quantification techniques 

• Other high-performance applications 

Architecture and Networks: All aspects of high-performance hardware including the optimization and evaluation of processors and networks. 

Topics include: 

• Innovative hardware/software co-design 

• Interconnect technologies (e.g., InfiniBand, Myrinet, Ethernet and Routable PCI), switch/router architecture, network topologies, on-chip or optical networks and network fault tolerance 

• Software defined networks 

• Memory systems, novel memory architectures, caches 

• Parallel and scalable system architectures 

• Power-efficient, resilient, highly-available, stream, vector, embedded and reconfigurable architectures, and emerging technologies 

• Processor architecture, chip multiprocessors, GPUs, custom and reconfigurable logic 

• Protocols (e.g., TCP, UDP and sockets), quality of service, congestion management and collective communication 

Clouds and Distributed Computing: All software aspects of clouds and distributed computing that are related to high-performance computing systems, including software architecture, configuration, optimization and evaluation. 

Topics include: 

• Compute and storage cloud architectures 

• Data management 

• Problem-solving environments 

• Programming models and tools 

• Management of Quality of service and service-level agreements 

• Scheduling, load balancing, resource provisioning, and energy efficiency 

• Self-configuration, management, information services and monitoring 

• Service-oriented architectures and tools for integration of clouds, clusters and distributed computing 

• Multitenancy, virtualization, and overlays 

• Security and identity management 

Data Analytics, Visualization and Storage: All aspects of data analytics, visualization and storage related to high-performance computing systems. 

Topics include: 

• Databases and scalable structured storage for HPC 

• Data mining, analysis and visualization for modeling and simulation 

• Ensemble analysis and visualization 

• I/O performance tuning, benchmarking and middleware 

• Scalable storage, next-generation storage systems and media 

• Parallel file, storage and archival systems 

• Provenance, metadata and data management 

• Reliability and fault tolerance in HPC storage 

• Scalable storage, metadata and data management 

• Storage networks 

• Storage systems for data intensive computing 

• Data science 

• Visualization and image processing 

Performance Measurement, Modeling, and Tools: Novel methods and tools for measuring, evaluating, and/or analyzing performance. “Performance” may be broadly construed to include any number of metrics, such as execution time, energy, power, or potential measures of resilience. Submissions in this area are encouraged to show the applicability and reproducibility of their results by means such as sensitivity analysis, performance modeling, or code snippets.

Topics include: 

• Analysis, modeling, or simulation methods 

• Empirical measurement techniques on real-world systems 

• Scalable tools and instrumentation infrastructure for measurement, monitoring, and/or visualization of performance 

• Novel, broadly applicable performance optimization techniques 

• Methodologies, metrics, and formalisms for performance analysis and tools 

• Performance studies of HPC subsystems, such as processor, network, memory and I/O 

• Workload characterization and benchmarking techniques 

Programming Systems: Technologies that support parallel programming for large-scale systems as well as smaller-scale components that will plausibly serve as building blocks for next-generation high-performance computing architectures. 

Topics include: 

• Programming language techniques for reducing energy and data movement (e.g., precision allocation, use of approximations, tiling) 

• Solutions for parallel programming challenges (e.g., interoperability, memory consistency, determinism, race detection, work stealing or load balancing) 

• Parallel application frameworks 

• Tools for parallel program development (e.g., debuggers and integrated development environments) 

• Program analysis, synthesis, and verification to enhance cross-platform portability, maintainability, result reproducibility, resilience (e.g., combined static and dynamic analysis methods, testing, formal methods) 

• Compiler analysis and optimization; program transformation 

• Parallel programming languages, libraries, models and notations 

• Runtime systems as they interact with programming systems 

State of the Practice: All aspects related to novel but at the same time pragmatic practices of HPC that allow for results that are far superior with respect to time-, energy-, or cost-to-solution. These include infrastructure, services, facilities and large-scale application executions. Submissions that develop best end-to-end practices, optimized designs or benchmarks are of particular interest. Although concrete case studies within a conceptual framework often serve as the basis for accepted papers, how the experience generalizes is particularly encouraged. 

Topics include: 

• Bridging of cloud data centers and supercomputing centers 

• Comparative system benchmarking over a wide spectrum of workloads 

• Deployment experiences of large-scale infrastructures and facilities 

• Facilitation of “big data” associated with supercomputing 

• Long-term infrastructural management experiences 

• Pragmatic resource management strategies and experiences 

• Procurement, technology investment and acquisition best practices 

• Quantitative results of education, training and dissemination activities 

• User support experiences with large-scale and novel machines 

• Infrastructural policy issues, especially international experiences 

• Software engineering best practices for HPC 

System Software: Operating system (OS), runtime system and other low-level software research & development that enables allocation and management of hardware resources for high-performance computing applications and services. 

Topics include: 

• Alternative and specialized parallel operating systems and runtime systems 

• Approaches for enabling adaptive and introspective system software 

• Communication optimization 

• Distributed shared memory systems 

• System support for global address spaces 

• Enhancements for attached and integrated accelerators 

• Interactions between the OS, runtime, compiler, middleware, and tools 

• Parallel/networked file system integration with the OS and runtime 

• Resource management 

• Run-time and OS management of complex memory hierarchies 

• System software strategies for controlling energy and temperature 

• Support for fault tolerance and resilience 

• Virtualization and virtual machines