The performance of μ-kernel-based systems

@article{Hrtig1997ThePO,
  title={The performance of $\mu$-kernel-based systems},
  author={Hermann H{\"a}rtig and Michael Hohmuth and Jochen Liedtke and Sebastian Sch{\"o}nberg and Jean Wolter},
  journal={Proceedings of the sixteenth ACM symposium on Operating systems principles},
  year={1997}
}
First-generation μ-kernels have a reputation for being too slow and lacking sufficient flexibility. To determine whether L4, a lean second-generation μ-kernel, has overcome these limitations, we have repeated several earlier experiments and conducted some novel ones. Moreover, we ported the Linux operating system to run on top of the L4 μ-kernel and compared the resulting system with both Linux running native, and MkLinux, a Linux version that executes on top of a first-generation Mach-derived… 
The Cost of IPC : an Architectural Analysis
TLDR
The work shows the increasing architecture cost of these operations, especially in the privilege mode changes and address space switches, as a big challenge for future processors to deliver high performance and fault-tolerant secure systems based on a μ-kernel design.
L4/Darwin: Evolving UNIX
TLDR
The on-going design and implementation of the Darbat project and the experiences of bringing the strengths of UNIX into the UNIX kernel itself are covered.
Hijack COSLINUX : Toward Practical , Predictable , and Efficient OS Co-Location using Linux ∗
TLDR
An OS design that enables the co-location of both the Linux kernel, and other real-time operating system kernels on the same system, and the use of this technique to co-locate with Linux the COMPOSITE component-based OS is outlined.
A lightweight virtual machine monitor for Blue Gene/P
TLDR
This paper presents a lightweight, micro-kernel-based virtual machine monitor (VMM) for the Blue Gene/P supercomputer, providing the convenience of a fully featured commodity software stack, while also promising to deliver the scalability and low latency of an HPC operating system.
Composition Kernel: A Software Solution for Constructing a Multi-OS Embedded System
TLDR
A method called a composition kernel is introduced which uses a thin abstraction layer for accommodating kernels without making direct dependencies between them and was significantly smaller than that in related work.
KLOS : A High Performance Kernel-Less Operating System
TLDR
It is discovered that it is possible to achieve a significant performance gain in accessing the services provided by the operating system by not performing a ring transition, and can be achieved without compromising on the separation of the privileged components from the non-privileged.
Providing Protected Execution Environments for Embedded Operating Systems Using a µ-Kernel
TLDR
This paper presents an approach, in which a microkernel is employed to provide protected execution environments for the existing embedded kernels, and develops the system that consists of TL4 microkernel and a μITRON embedded kernel, and shows the feasibility of the approach through evaluations.
Using the OSKit as a base for L4 applications
TLDR
This paper describes the port of the OSKit to L4, and shows ways of using the OS Kit and assess their difficulty, and aims at making L4 developers more accustomed with the OSkit.
A High Performance Kernel-Less Operating System Architecture
TLDR
KLOS is a Kernel-Less Operating System built on the basis of such a design and the KLOS service call mechanism is an order of magnitude faster than the current widely implemented mechanisms for service or system calls with a 4x improvement over the traditional trap/interrupt and a 2x improved over the Intel SYSENTER/SYSEXIT fast system call models.
Think: A Software Framework for Component-based Operating System Kernels
TLDR
Performance measurements show no degradation due to componentization and the systematic use of the binding framework, and that application-specific kernels can achieve speed-ups over standard general-purpose operating systems such as Linux.
...
...

References

SHOWING 1-10 OF 110 REFERENCES
On micro-kernel construction
TLDR
Contradictory to this belief, it is shown and support by documentary evidence that inefficiency and inflexibility of current μ-kernels is not inherited from the basic idea but mostly from overloading the kernel and/or from improper implementation.
On -Kernel Construction
TLDR
It is shown and support by documentary evidence that ine ciency and in exibility of current -kernels is not inherited from the basic idea but mostly from overloading the kernel and/or from improper implementation.
The impact of operating system structure on memory system performance
TLDR
This paper evaluates the memory system behavior of two distinctly different implementations of the UNIX operating system: DEC's Ultrix, a monolithic system, and Mach 3.0 with CMU's UNIX server, a microkernel-based system, showing that the micro kernel- based system executes substantially more non-idle system instructions for an equivalent workload than themonolithic system.
Improving IPC by kernel design
TLDR
The main ideas are to guide the complete kernel design by the ipc requirements, and to make heavy use of the concept of virtual address space inside the μ-kernel itself.
Linux on the OSF Mach3 microkernel
TLDR
The architecture of the Linux server is described, emphasizing the areas requiring interaction between the Linux code and the microkernel, and some performance improvements are presented.
Toward real microkernels
TLDR
The second-generation microkernels may be a basis for all types of operating systems, including timesharing, mul-timedia, and soft and hard real time.
Dealing with disaster: surviving misbehaved kernel extensions
TLDR
This paper explains how VINO uses software fault isolation as its safety mechanism and a lightweight transaction system to cope with resource-hoarding and finds that while the overhead of these techniques is high relative to the cost of the extensions themselves, it is lowrelative to the benefits that extensibility brings.
Operating system benchmarking in the wake of lmbench: a case study of the performance of NetBSD on the Intel x86 architecture
TLDR
Modifications to lmbench are described, and a new benchmark suite, hbench:OS, is used to examine how the performance of operating system primitives under NetBSD has scaled with the processor evolution of the Intel x86 architecture.
Optimistic incremental specialization: streamlining a commercial operating system
TLDR
The ideas outlined in this paper allow the conventional kernel design tenet of “optimi zing for the common case” to be extended to the domain of adaptive operating systems and it is shown that aggressive use of specialization can produce in-kernel implementations of operating system functionality with performance comparable to user-level implementations.
Generic virtual memory management for operating system kernels
TLDR
The rationale and design of a Generic Memory management Interface is discussed, for a family of scalable operating systems, which provides abstractions for support of a single, consistent cache for both mapped objects and explicit I/O, and control of data caching in real memory.
...
...