Towards effective user-controlled scheduling for microkernel-based systems

Abstract

With μ-kernel based systems becoming more and more prevalent, the demand for extensible resource management raises - and with it the demand for flexible thread scheduling. In this paper, we investigate the benefits and costs of a μ-kernel that exports scheduling from the kernel to user level. A key idea of our approach is to involve the user level whenever the μ-kernel encounters a situation that is ambiguous with respect to scheduling, and to permit the kernel to resolve the ambiguity based on user decisions. A further key aspect is that we rely on a generic, protection domain neutral interface between kernel and applications. For evaluation, we have developed a hierarchical user level scheduling architecture for the L4 μ-kernel, and a virtualization environment running on its top. Our environment supports Linux 2.6.9 guest operating systems on IA-32 processors. Experiments indicate an application overhead between 0 and 10 percent compared to a pure in-kernel scheduler solution, but also demonstrate that our architecture enables effective and accurate user-directed scheduling.

Predictable Interrupt Management and Scheduling in the Composite Component-Based System

“…Currently, this operation is of comparable efficiency to other systems with a focus on IPC efficiency such as L4 [30]. We believe that optimizing the fast-path in Composite by writing it in assembly can further reduce latency.…”

“…Unlike previous systems that provide user-level scheduling [11,4,30], the operation of blocking in Composite is not built into the underlying kernel. This means that schedulers are able to provide customizable blocking semantics.…”

“…Past research has put forth mechanisms to implement hierarchically structured user-level schedulers [11,30].…”

HiRes: A System for Predictable Hierarchical Resource Management

“…Some form the hierarchy of schedulers in the kernel, requiring that they be trusted [19], [20]. Others use CPU donations [21], [22] to enable any thread to grant processing to any other. Though general, this makes it difficult for a parent to control the assignment of all resources to a child.…”