Click is a new software architecture for building flexible and configurable routers. A Click router is assembled from packet processing modules called elements. Individual elements implement simple router functions like packet classification, queueing, scheduling, and interfacing with network devices. A router configuration is a directed graph with elements at the vertices; packets flow along the edges of the graph. Configurations are written in a declarative language that supports user-defined abstractions. This language is both readable by humans and easily manipulated by tools. We present language tools that optimize router configurations and ensure they satisfy simple invariants. Due to Click’s architecture and language, Click router configurations are modular and easy to extend. A standards-compliant Click IP router has sixteen elements on its forwarding path. We present extensions to this router that support dropping policies, fairness among flows, quality-of-service, and

We describe an operating system architecture that securely multiplexes machine resources while permitting an unprecedented degree of application-specific customization of traditional operating system abstractions. By abstracting physical hardware resources, traditional operating systems have significantly limited the performance, flexibility, and functionality of applications. The exokernel architecture removes these limitations by allowing untrusted software to implement traditional operating system abstractions entirely at application-level. We have implemented a prototype exokernel-based system that includes Aegis, an exokernel, and ExOS, an untrusted application-level operating system. Aegis defines the low-level interface to machine resources. Applications can allocate and use machine resources, efficiently handle events, and participate in resource revocation. Measurements show that most primitive Aegis operations are 10–100 times faster than Ultrix,a mature monolithic UNIX operating system. ExOS implements processes, virtual memory, and inter-process communication abstractions entirely within a library. Measurements show that ExOS’s application-level virtual memory and IPC primitives are 5–50 times faster than Ultrix’s primitives. These results demonstrate that the exokernel operating system design is practical and offers an excellent combination of performance and flexibility. 1

The need for supportingvariety of hard and soft real-time, as well as best effort applications in a multimedia computing environment requires an operating system framework that: (1) enables different schedulers to be employed for different application classes, and (2) provides protection between the various classes of applications. We argue that these objectives can be achieved by hierarchical partitioning of CPU bandwidth, in which an operating system partitions the CPU bandwidth among various application classes, and each application class, in turn, partitions its allocation (potentially using a different scheduling algorithm) among its sub-classes or applications. We present Start-time Fair Queuing (SFQ) algorithm, which enables such hierarchical partitioning. We have implemented a hierarchical scheduler in Solaris 2.4. We describe our implementation, and demonstrate its suitability for multimedia operating systems. 1

Most operating systems use interface interrupts to schedule network tasks. Interrupt-driven systems can provide low overhead and good latency at low of-fered load, but degrade significantly at higher arrival rates unless care is taken to prevent several pathologies. These are various forms of receive livelock, in which the system spends all its time processing interrupts, to the exclusion of other neces-sary tasks. Under extreme conditions, no packets are delivered to the user application or the output of the system. To avoid livelock and related problems, an operat-ing system must schedule network interrupt handling as carefully as it schedules process execution. We modified an interrupt-driven networking implemen-tation to do so; this eliminates receive livelock without degrading other aspects of system performance. We present measurements demonstrating the success of our approach. 1.

We propose and analyze a proportional share resource allocation algorithm for realizing real-time performance in time-shared operating systems. In a proportional share system, processes are assigned a weight which determines a share (percentage) of the resource they are to receive. The resource is then allocated in discrete-sized time quanta in such a manner that each process makes progress at a precise, uniform rate. Proportional share allocation algorithms are of interest because (1) they provide a natural means of seamlessly integrating real- and non-real-time processing requirements in a general purpose operating system, (2) they are easy to implement (and in particular, easier than more traditional forms of real-time support such as periodic tasks), (3) they provide a simple and effective means of precisely controlling the real-time performance of a process including uniform, predictable degradation in times of system overload, and (4) they provide a natural mean of policing proce...

This thesis presents flexible abstractions for specifying resource management policies, together with efficient mechanisms for implementing those abstractions. Several novel scheduling techniques are introduced, including both randomized and deterministic algorithms that provide proportional-share control over resource consumption rates. Such control is beyond the capabilities of conventional schedulers, and is desirable across a broad spectrum of systems that service clients of varying importance. Proportional-share scheduling is examined for several diverse resources, including processor time, memory, access to locks, and disk bandwidth. Resource rights are encapsulated by abstract, first-class objects called tickets. An active client consumes resources at a rate proportional to the number of tickets that it holds. Tickets can be issued in different amounts and may be transferred between clients. A modular currency abstraction is also introduced to flexibly name, share, and protect ...

Disk schedulers in current operating systems are generally work-conserving, i.e., they schedule a request as son as the previous request has finished. Such schedulers often require multiple outstanding requests from each process to meet system-level goals of performance and quality of service. Unfortunately, many common applications issue disk read requests in a synchronous manna% interspersing successive requests with shor periods of computation. The scheduler chooses the next request too early; this induces deceptive idleness, a condition where the scheduler incorrectly assumes that the test request issuing process has no further requests, and becomes forced to switch to a toques? from another pro- Ce3S.

\Soft state " is an often cited yet vague concept in network protocol design in which two or more network entities intercommunicate in a loosely coupled, often anonymous fashion. Researchers often de ne this concept operationally (if at all) rather than analytically: a source of soft state transmits periodic \refresh messages " over a (lossy) communication channel to one or more receivers that maintain a copy of that state, which in turn \expires " if the periodic updates cease. Though a number of crucial Internet protocol building blocks are rooted in soft state-based designs | e.g., RSVP refresh messages, PIM membership updates, various routing protocol updates, RTCP control messages, directory services like SAP, and so forth | controversy is building as to whether the performance overhead of soft state refresh messages justify their qualitative bene t of enhanced system \robustness". We believe that this controversy has risen not from fundamental performance tradeo s but rather from our lack of a comprehensive understanding of soft state. To better understand these tradeo s, we propose herein a formal model for soft state communication based on a probabilistic delivery model with relaxed reliability. Using this model, we conduct queueing analysis and simulation to characterize the data consistency and performance tradeo s under a range of workloads and network loss rates. We then extend our model with feedback and show, through simulation, that adding feedback dramatically improves data consistency (by up to 55%) without increasing network resource consumption. Our model not only provides a foundation for understanding soft state, but also induces a new fundamental transport protocol based on probabilistic delivery. Toward this end, we sketch our design of the \Soft State Transport Protocol " (SSTP), which enjoys the robustness of soft state while retaining the performance bene t of hard state protocols like TCP through its judicious use of feedback. This research was supported by DARPA contract N66001-96-C-8508, by the State of California under the MICRO program, and by

In this paper, we introduce a new operating system abstraction called reservation domains, and describe its implementation in Eclipse, an experimental operating system that provides a testbed for Quality of Service (QoS) support for applications. Reservation domains enable explicit control over the provisioning of system resources among applications in order to achieve desired levels of predictable performance. In general, each reservation domain is assigned a certain fraction of each resource (e.g., 25% CPU, 50% disk I/O, etc.). Eclipse implements reservation-domain scheduling of multiple resources. It currently supports CPU and disk and physical memory (working set size) scheduling. Eclipse implements a new scheduling algorithm, Move-to-Rear List Scheduling (MTR-LS), that provides a cumulative service guarantee, in addition to fairness and delay bounds. Cumulative service guarantee is necessary for ensuring predictable aggregate throughput for applications that require multiple res...

Traditional processor scheduling mechanisms in operating systems are fairly rigid, often supportingonly one fixed scheduling policy, or, at most, a few "scheduling classes" whose implementations are closely tied together in the OS kernel. This paper presents CPU inheritance scheduling, a novel processor scheduling framework in which arbitrary threads can act as schedulers for other threads. Widely different scheduling policies can be implemented under the framework, and many different policies can coexist in a single system, providing much greater scheduling flexibility. Modular, hierarchical control can be provided over the processor utilization of arbitrary administrative domains, such as processes, jobs, users, and groups, and the CPU resources consumed can be accounted for and attributed accurately. Applications, as well as the OS, can implement customized local scheduling policies; the framework ensures that all the different policies work together logically and predictably. As a ...