Topic 4. High Performance Architectures and Compilers
This topic deals with architecture design, languages, and compilation for parallel high performance systems. The areas of interest range from microprocessors to large-scale parallel machines (including multi-/many-core, possibly heterogeneous, architectures); from general-purpose to specialized hardware platforms (e.g., graphic coprocessors, low-power embedded systems); and from architecture design to compiler technology and language design.
On the compilation side, topics of interest include programmer productivity issues, concurrent and/or sequential language aspects, vectorization, program analysis, program transformation, automatic discovery and/or management of parallelism at all levels, autotuning and feedback directed compilation, and the interaction between the compiler and the system at large. On the architecture side, the scope spans system architectures, processor micro-architecture, memory hierarchy, and multi-threading, architectural support for parallelism, and the impact of emerging hardware technologies.
Focus
- Compiling for multi-threaded/multi-core/many-core/vector and heterogeneous processors/architectures
- Compiling for emerging architectures (low-power embedded systems, reconfigurable hardware, processors in memory, coprocessors)
- Iterative, just-in-time, feedback-oriented, dynamic, and machine-learning-based compilation
- Static analysis and interaction between static and dynamic analysis
- Programmer productivity tools and analysis for high-performance architectures
- Program transformation systems
- High level programming models and tools for multi-/many-core and heterogeneous architectures
- Interaction between compiler, runtime system, application, hardware, and operating system
- Parallel computer architecture design – ILP, DLP, multi-threaded, and multi-core processors
- Designs and compiler optimizations for power/performance efficiency
- Software and hardware fault-tolerance techniques
- Memory hierarchy, emerging memory technologies, and 3D stacked memories
- Application-specific, reconfigurable and embedded parallel systems
- Compiler, run-time, and architectural support for dynamic adaptation
- Optimizing compilers for Domain Specific Languages
Chairs
- Global chair
- Florian Brandner, Télécom ParisTech, Université Paris-Saclay, France
- Local chair
- Fabio Luporini, Imperial College London, UK
- Chair
- Alexandra Jimborean, Uppsala University, Sweden
- Chair
- Frank Hannig, Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany
- Chair
- Gihan Mudalige, University of Warwick, UK