Virtual Open Systems (France)

Embedded systems are pervasively used in many fields nowadays. In mixed-criticality environments (automotive, industry 4.0, drones, etc.) they need to run real-time applications with certain time and safety constraints alongside a rich operating system (OS). This is usually possible thanks to virtualization techniques, that leverage on hardware virtualization extensions on the machine. However, these hardware extensions might not cope with the security and safety requirements of the specific use case, and additionally, they might not always be available. A notable example is the emerging RISC-V architecture, that is today gaining a lot of traction in the mixed criticality field, but that does not offer today hardware virtualization extensions. In this paper VOSySmonitoRV is proposed as a mixed-criticality solution for RISC-V systems. VOSySmonitoRV allows the co-execution of two or more operating systems in a secure and isolated manner by running in the highest privileged machine level. A specific benchmark, measuring the interrupt latency and context switch time is done to assess the system performance in mixed criticality systems.

Smart cities are one of the most important 5G verticals due to their impact on people's life. The neutral host model is key for this vertical, specially to promote the infrastructure sharing between operators for a pervasive infrastructure deployment. However, it demands enhancements in today's virtualization technologies to support geographically scattered and resource constrained computing and networking elements (i.e., core data centers, edge computing point of presences, and far-edge servers at lampposts, etc.). This paper presents the virtualization enhancements developed within the H2020 5GCity project to support the neutral host model for smart city deployments. The envisioned virtualization technology demands security and technology extensions with a unified view of sliceable and heterogeneous devices and radio technologies (e.g., LTE, 5G, and Wi-Fi). In our solution, security is enhanced by deploying a robust compute node authentication, monitoring, and geo-tagging. In addition, wireless connectivity is extended via an innovative multiple RAN controller approach for the management and control of heterogeneous radio resources. In order to be validated, our virtualization approach is being deployed and demonstrated in the cities of Bristol, Barcelona, and Lucca.

Nowadays cloud-computing systems take a great advantage of virtualization for the benefits of workload isolation and flexible resources partitioning. It is expected that the same functionalities will be available also on disaggregated architectures, proposed recently as next generation approach for building data-centers. In this publication, we are presenting the design and prototype of an enhanced virtualization layer, enabling runtime memory balancing between virtual machines on a section granularity. Guests' RAM is backed by isolated chunks of host memory, coming from independent physical banks, not necessarily a local one. It can be dynamically resized without requiring any support for the ACPI emulation in the virtualization framework, as we exemplified by implementing the prototype on ARMv8 platform.

Platform-As-A-Service (PaaS) systems offer customers a rich environment in which to build, deploy, and run applications. Today's PaaS offerings are tailored mainly to the needs of web and mobile applications developers, and involve a fairly rigid stack of components and features. The vision of the H2020 5GPPP Phase 2 Next Generation Platform-as-a-Service (NGPaaS) project is to enable "build-to-order" customized PaaSes, tailored to the needs of a wide range of use cases with telco-grade 5G characteristics. This paper sets out the salient and innovative features of NGPaaS and explores the impacts on Operational Support Systems and Business Support Systems (OSS/BSS), moving from fixed centralized stacks to a much more flexible and modular distributed architecture.

Integrating multiple subsystems with different levels of criticality is a well established concept in the automotive domain. To ensure proper temporal and spatial isolation, a highly privileged software component is installed to orchestrate the subsystems. VOSYSmonitor is such a solution, it enables the co-execution of two operating systems on a single System on Chip - A rich operating system, such as Linux, along with a safety critical operating system, fully isolated from each other using ARM TrustZone. But if we take a closer look at specific automotive scenarios (e.g., “displaying warning signs”), reveals that an interaction of the two operating systems might be desirable. In this paper we address this challenge. We present the implementation of a low-latency inter-world network channel. It is built around already existing primitives in both worlds, only implementing the physical layer of the network channel. This ensures a low complexity, meaning only minor modifications have to be made to both operating systems. To prove the feasibility of our design, we built a full prototype that enables a network communication between the two operating systems, while ensuring a proper encapsulation of the safety critical operating system. To validate low reaction times, the design is evaluated with respect to network latency. To complement the measurements, we also performed a number of bandwidth measurements. Finally, we thoroughly discuss potential threat scenarios arising from the network link and how they can be addressed with appropriate countermeasures.

Network Functions Virtualisation (NFV) is revolutionizing the telecom operators data centers as well as the way in which virtualization is applied to high performance/high availability scenarios. Hardware accelerators, and in particular FPGAs, are a very promising solution for NFV, both for their high performance, predictability, programmability and reconfigurability. Recently, consolidation is brought to its limit in this field with lightweight virtualization techniques, which require a high overcommitment of FPGA accelerators resources in order to cope with the needs of thousands of guests. The lack of such virtualization support, together with cloud integration and programmability issues, limits the deployment of FPGAs in NFV. This paper presents the Virtualization Manager (VirtManager), an FPGA IP block which enables accelerators overcommitment with direct guest access. This is a key feature to enable the acceleration of thousands of guests, as it is the case for lightweight VMs, containers and unikernels. This paper's main contribution is a novel architecture for FPGA virtualization with support for resources overcommitment, NFV management and orchestration. A performance analysis of the accelerators context switch mechanism is included to prove the feasibility of the proposed approach.

Disaggregation and rack-scale systems have the potential of drastically increasing TCO and utilization of cloud datacenters, while maintaining performance. In this paper, we present a novel rack-scale system architecture featuring software-defined remote memory disaggregation. Our hardware design and operating system extensions enable unmodified applications to dynamically attach to memory segments residing on physically remote memory pools and use such remote segments in a byte-addressable manner, as if they were local to the application. Our system features also a control plane that automates software-defined dynamic matching of compute to memory resources, as driven by datacenter workload needs. We prototyped our system on the commercially available Zynq Ultrascale+ MPSoC platform. To our knowledge, this is the first time a software-defined disaggregated system has been prototyped on commercial hardware and evaluated through industry standard software benchmarks. Our initial results — using benchmarks that are artificially highly adversarial in terms of memory bandwidth — show that disaggregated memory access exhibits a round-trip latency of only 134 clock cycles; and a throughput penalty of as low as 55%, relative to locally-attached memory. We also discuss estimations as to how our findings may translate to applications with pragmatically milder memory aggressiveness levels, as well as innovation avenues across the stack opened up by our work.

Remote DMA (RDMA) engines are widely used in clusters/data-centres to improve the performance of data transfers between applications running on different nodes of a computing system. RDMAs are today supported by most network architectures and distributed programming models. However, with the massive usage of virtualization most applications will use RDMAs from virtual machines, and the virtualization of such I/O devices poses several challenges. This paper describes a generic para-virtualization framework based on API Remoting, providing at the same time the flexibility of software based virtualization, and the low overhead of hardware-assisted solutions. The solution presented in this paper is targeting the KVM hypervisor, but is not bound to any target network architecture or specific RDMA engine, thanks to the virtualization at the level of the programming API. In addition, two of the major limitations of para-virtualization are addressed: data sharing between host and guest, and interactions between guests and hypervisor. A set of experimental results showed a near to native performance for the final user of the RDMA (i.e., maximum transfer bandwidth), with a higher overhead only to simulate the API functions used to initialize the RDMA device or allocate/deallocate RDMA buffers.

Cyber-physical systems (CPS) are devices with sensors and actuators which link the physical with the virtual world. For CPS there is a strong trend towards open systems, which can be extended during operation by instantly adding functionalities on demand. The main goal of the TAPPS (Trusted Apps for open CPS) project is the development of a platform for CPS apps which can also access and modify safety critical device internals. As current, rich execution platforms for apps are limited in security, TAPPS will provide and validate an end-to-end solution for development and deployment of trusted apps. The project will develop a dedicated, real-time Trusted Execution Environment (TEE) for highly-trusted CPS apps. Additionally, TAPPS also includes an App Store and a model-based tool chain for trusted application development including verification tools. The multi-level trusted apps platform and tool chain are matured and validated in health and automotive application domains using industrial, realistic use cases paving the way for future exploitation in further demanding application domains.

VAS : An Opera in Flatland (2002) de Steve Tomasula a été salué par la critique comme l’une des œuvres littéraires les plus novatrices et les plus subtiles dans sa combinaison du texte et de l’image. The Book of Portraiture (2006) peut, au premier regard, sembler moins audacieux. La question au cœur de l’œuvre est celle du portrait – ou du désir plus général de l’humanité de donner des représentations d’elle-même sous forme de tableaux, de rêves, de photographies, de données électroniques et de séquençage ADN – question qui se déploie de chapitre en chapitre, reliés par une narration complexe. Tandis que les images occupent une place centrale dans la diégèse, elles semblent aussi constituer un fond sous-exposé au texte écrit. Cet article se veut une démonstration que ces éléments graphiques, de la couleur des pages aux cadrage du texte, font partie intégrale du projet de Tomasula, et que leur portée dépasse de loin le décoratif ou l’ornemental. Il faut y voir une « strate » du récit, dont la véritable signification ne peut être perçue que lorsque les lecteurs font la mise au point non plus sur le texte surexposé, mais sur cet arrière-plan visuel.