Alexandre Courbot

Hardware-Assisted Reliability Enhancement for Embedded Multi-core Virtualization Design

In this paper, we propose a virtualization architecture for the multi-core embedded system to provide more system reliability and security while maintaining the same performance without introducing additional special hardware supports or having to implement complex protection mechanism in the virtualization layer. Virtualization has been widely used in embedded systems, especially in consumer electronics, albeit itself is not a new technique, because there are various needs for both GPOS (General Purpose Operating System) and RTOS (Real Time Operating System). The surge of the multi-core platform in the embedded system also helps the consolidation of the virtualization system for its better performance and lower power consumption. Embedded virtualization design usually uses two kinds of approaches. The first one is to use the traditional VMM, but it is too complicated for use in the embedded environment if there is no additional special hardware support. The other is the use of the micro kernel which imposes a modular design. The guest systems, however, would suffer from considerable amount of modifications because the micro kernel lets the guest systems to run in user space. For some RTOSes and theirs applications originally running in kernel space, it makes this approach more difficult to work because a lot of privileged instructions are used in those codes. To achieve better reliability and keep the virtualization layer design light weighted, a common hardware component adopted in the multi-core embedded processors is used in this work. In the most embedded platforms, vendors provide additional on-chip local memory for each physical core and these local memory areas are private only to their cores. By taking this memory architectures advantage, we can mitigate above-mentioned problems at once. We choose to re-map the virtualization layers program called SPUMONE, which it runs all its guest systems in kernel space, on the local memory. By doing so, it can provide additional reliability and security for the entire system because the SPUMONEs design in a multi-core platform has each instance being installed on a separated processor core which is different from the traditional virtualization layer design and the content of each SPUMONE is inaccessible to each others. We also achieve this goal without bringing any overhead to the overall performance.

Temporal and spatial isolation in a virtualization layer for multi-core processor based information appliances

A virtualization layer makes it possible to compose multiple functionalities on a multi-core processor with minimum modifications of OS kernels and applications. A multi-core processor is a good candidate to compose various software independently developed for dedicated processors into one multi-core processor to reduce both the hardware and development cost. In this paper, we present SPUMONE, which is a virtualization layer suitable for developing multi-core processor based-information appliances.

Composition kernel: a multi-core processor virtualization layer for rich functional smart products

Future ambient intelligence environments will embed powerful multi-core processors to compose various functionalities into a smaller number of hardware components. This makes the maintainability of intelligent environments better because it is not easy to manage massively distributed processors. A composition kernel makes it possible to compose multiple functionalities on a multi-core processor with the minimum modification of OS kernels and applications. A multi-core processor is a good candidate to compose various software developed independently for dedicated processors into one multi-core processor to reduce both the hardware and development cost. In this paper, we present SPUMONE which is a composition kernel for developing future smart products.

A Lightweight Monitoring Service for Multi-core Embedded Systems

The recent increase in complexity and functionality in embedded systems makes them more vulnerable to rootkit-type attacks, raising the need for integrity management systems. However, as of today there is no such system that can guarantee the system’s safety while matching the low-resource, real-time and multi-core requirements of embedded systems. In this paper, we present a Virtual Machine Monitor (VMM)based monitoring service for embedded systems that checks the actual kernel data against a safe data specification. However, due to the VMM and multi-core nature of the system, the guest OS can be preempted at any time, leading to the checking of potentially inconsistent states. We evaluated two approaches to solve this problem: detecting such invalid states by checking specific kernel data, and detecting system calls using the VMM.

A Framework for Self-Healing Device Drivers

Device drivers are the major cause of operating system failure. Prior research proposed frameworks to improve the reliability of device drivers by means of driver restart. While avoiding any instrumentation of the driver, this approach does not always allow graceful recovery. In this paper, we propose a framework for self-healing device drivers that lets the driver developer consider and implement the failure recovery of device drivers. For this purpose, our framework provides easy to use and light-weight persistent memory that preserves the state of the driver needed to successfully recover. We developed a prototype on top of the L4 microkernel, and were able to achieve full recovery of crashed drivers as fast as 0.2 ms for different device drivers. In all cases, recovery was totally transparent for the user.

An Operating System Architecture for Future Information Appliances

A software platform for developing future information appliances requires to satisfy various diverse requirements. The operating system architecture presented in this paper enhances the flexibility and dependability through virtualization techniques. The architecture allows a system to use multiple operating systems simultaneously, and to use multi-core processors in a flexible way. Also, dependability mechanisms in our architecture will avoid crashing or hanging a system as much as possible in order to improve the user experience when defects in the software are exposed. We present a brief overview of each component in the operating system architecture and some sample scenarios that illustrate the effectiveness of the architecture.

mBrace: action-based performance monitoring of multi-tier web applications

Today’s monitoring tools and web application middleware offer very limited support for detailed performance analysis of HTTP request-based, multi-tier web applications. They measure resource usage system-wide or per command name, but are unable to isolate usage per action specific to the target application1. This inability has several causes: First, many HTTP requests are executed by one application or script simultaneously, making it hard to isolate them. Second, it is difficult to monitor at the granularity of a thread, while efficient web servers and database servers implement thread-based process models. Third, different types of actions are often handled by one application or script, hindering type-based distinction. Finally, existing tools are often interval-based, consequently missing short-lived processes that live between two capturing points. The same limitations apply to measuring individual SQL queries. In addition, SQL queries cannot be backtracked to the HTTP request that triggered them, even though they frequently turn out to be the cause of long end-to-end response times.This paper describes a preliminary solution that addresses these problems and enables us to measure resource consumption and database interaction for individual actions, in real-time and under load, with minimal effort from the web application developer.

Coexisting real-time OS and general purpose OS on an embedded virtualization layer for a multicore processor

Porting operating systems to a virtualization layer produces a semantic gap because the assumptions that guest OSes rely on may not be ensured. On multi-core environments, this gap can cause the fatal performance degradations. The lock holder preemption (LHP) problem is a well known example of the sources of the performance degradation. It occurs when a thread holding a spin lock in an OS kernel is preempted by other OS kernels. Some previous proposals can avoid this problem, but none of them cares about the real-time responsiveness of guest OSes. Therefore the approaches are not suitable for embedded systems. We have developed a new technique for avoiding the LHP problem. The approach can ensure both the real-time responsiveness of RTOS and the high throughput of GPOS that supports shared memory multi-processors. This paper introduces the basic approach of our new technique and its experimental results. The results show that our new technique can make RTOS and GPOS coexist without degrading the real-time latency and is suitable to be applied to modern high performance multi-core processor based real-time embedded systems.

Composition Kernel: A Multi-core Processor Virtualization Layer for Highly Functional Embedded Systems

In this paper, we propose a composition kernel where multiple OS kernels are running on top of a very thin hardware abstraction layer. A composition kernel can reduce the engineering cost of developing an embedded system by reusing existing OS kernels and application with minimum modification without assuming special hardware supports.

Design issues in composition kernels for highly functional embedded systems

Recent embedded systems are dramatically increasing their features to satisfy a users demanding requirements. In traditional approaches, each feature uses a dedicated processor to be implemented, but the approach increases the cost to develop highly functional embedded systems. A multi-core processor offers a new opportunity to solve the dilemma because various functionalities can be implemented on one multi-core processor. While the number of functionalities is increased, we can just add more cores in the multi-core processor. However, we need a desirable software platform to implement a large number of functionalities on a multi-core processor for developing highly functional embedded systems. A composition kernel makes it possible to compose multiple functionalities on a multi-core processor with the minimum modifications of OS kernels and applications. In this paper, we present SPUMONE, a composition kernel for developing highly functional embedded systems, and runs on the SH4-based multi-core processor. SPUMONE does not support the spatial isolation to isolate guest OSes not to sacrifice the performance due to its virtualization. The approach may sacrifice the reliability and security, so we also propose to add the monitoring service that runs in the outside of a target guest OS to enhance its reliability and security without relying on the spatial isolation.