Compilation and virtualization in the HiPEAC vision
Christian Bertin, Christophe Guillon, Koen De Bosschere
Abstract
Christian Bertin, Christophe Guillon, Koen De Bosschere
Abstract
This paper describes the HiPEAC vision of embedded virtualization as it has developed during two years of discussion among the members of the HiPEAC cluster on binary translation and virtualization. We start from system virtualization and process virtualization and we gradually develop a vision in which the two merge into one virtualization layer for embedded systems. Such a unified virtualization offers solutions for consolidation, performance optimization, software engineering and dealing with legacy hardware components. Four adoption requirements are identified: support for real-time execution, low performance overhead, virtualization of accelerator cores and finally trustworthiness. Finally, we define four research challenges: full virtualization of heterogeneous multi-core platforms, portable performance for heterogeneous multi-cores, virtual machine management interfaces, and standards for embedded virtualization.
OpenAlex reports 6 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This paper describes the HiPEAC vision of embedded virtualization as it has developed during two years of discussion among the members of the HiPEAC cluster on binary translation and virtualization. We start from system virtualization and process virtualization and we gradually develop a vision in which the two merge into one virtualization layer for embedded systems. Such a unified virtualization offers solutions for consolidation, performance optimization, software engineering and dealing with legacy hardware components. Four adoption requirements are identified: support for real-time execution, low performance overhead, virtualization of accelerator cores and finally trustworthiness. Finally, we define four research challenges: full virtualization of heterogeneous multi-core platforms, portable performance for heterogeneous multi-cores, virtual machine management interfaces, and standards for embedded virtualization.
Key concepts: Full virtualization, Virtualization, Application virtualization, Hardware virtualization, Computer science, Service virtualization, Operating system, Virtual machine