Design and Implementation of Statistical Rate Monotonic Scheduling in KURT Linux

Statistical Rate Monotonic Scheduling (SRMS) is a generalization of the classical RMS results of Liu and Layland [LL73] for periodic tasks with highly variable execution times and statistical QoS requirements. The main tenet of SRMS is that the variability in task resource requirements could be smoothed through aggregation to yield guaranteed QoS. This aggregation is done over time for a given task and across multiple tasks for a given period of time. Similar to RMS, SRMS has two components: a feasibility test and a scheduling algorithm. SRMS feasibility test ensures that it is possible for a given periodic task set to share a given resource without violating any of the statistical QoS constraints imposed on each task in the set. The SRMS scheduling algorithm consists of two parts: a job admission controller and a scheduler. The SRMS scheduler is a simple, preemptive, fixed-priority scheduler. The SRMS job admission controller manages the QoS delivered to the various tasks through admit/reject and priority assignment decisions. In particular, it ensures the important property of task isolation, whereby tasks do not infringe on each other. In this paper we present the design and implementation of SRMS within the KURT Linux Operating System [HSPN98, SPH 98, Sri98]. KURT Linux supports conventional tasks as well as real-time tasks. It provides a mechanism for transitioning from normal Linux scheduling to a mixed scheduling of conventional and real-time tasks, and to a focused mode where only real-time tasks are scheduled. We overview the technical issues that we had to overcome in order to integrate SRMS into KURT Linux and present the API we have developed for scheduling periodic real-time tasks using SRMS.

Stinklepus Stinkleby Takes A Bath

This book was donated to BU by its author Stephanie Angelini, 2005 College of Communication alumna. All e-book versions of this work display best on the widest margin settings and the largest font setting. You may download the free Kindle reading software here: http://is.gd/yg9P5k ***PLEASE NOTE:*** The author has permitted us to make the book available to the BU community only. To download the files, please click on the appropriate lock icon and log in with your BU credentials. Thank you.

itmBench: Generalized API for Internet Traffic Managers

Internet Traffic Managers (ITMs) are special machines placed at strategic places in the Internet. itmBench is an interface that allows users (e.g. network managers, service providers, or experimental researchers) to register different traffic control functionalities to run on one ITM or an overlay of ITMs. Thus itmBench offers a tool that is extensible and powerful yet easy to maintain. ITM traffic control applications could be developed either using a kernel API so they run in kernel space, or using a user-space API so they run in user space. We demonstrate the flexibility of itmBench by showing the implementation of both a kernel module that provides a differentiated network service, and a user-space module that provides an overlay routing service. Our itmBench Linux-based prototype is free software and can be obtained from http://www.cs.bu.edu/groups/itm/.

Efficient End-Host Architecture for High Performance Communication Using User-level Sandboxing

Current low-level networking abstractions on modern operating systems are commonly implemented in the kernel to provide sufficient performance for general purpose applications. However, it is desirable for high performance applications to have more control over the networking subsystem to support optimizations for their specific needs. One approach is to allow networking services to be implemented at user-level. Unfortunately, this typically incurs costs due to scheduling overheads and unnecessary data copying via the kernel. In this paper, we describe a method to implement efficient application-specific network service extensions at user-level, that removes the cost of scheduling and provides protected access to lower-level system abstractions. We present a networking implementation that, with minor modifications to the Linux kernel, passes data between "sandboxed" extensions and the Ethernet device without copying or processing in the kernel. Using this mechanism, we put a customizable networking stack into a user-level sandbox and show how it can be used to efficiently process and forward data via proxies, or intermediate hosts, in the communication path of high performance data streams. Unlike other user-level networking implementations, our method makes no special hardware requirements to avoid unnecessary data copies. Results show that we achieve a substantial increase in throughput over comparable user-space methods using our networking stack implementation.

A Virtual Deadline Scheduler for Window-Constrained Service Guarantees

This paper presents a new approach to window-constrained scheduling, suitable for multimedia and weakly-hard real-time systems. We originally developed an algorithm, called Dynamic Window-Constrained Scheduling (DWCS), that attempts to guarantee no more than x out of y deadlines are missed for real-time jobs such as periodic CPU tasks, or delay-constrained packet streams. While DWCS is capable of generating a feasible window-constrained schedule that utilizes 100% of resources, it requires all jobs to have the same request periods (or intervals between successive service requests). We describe a new algorithm called Virtual Deadline Scheduling (VDS), that provides window-constrained service guarantees to jobs with potentially different request periods, while still maximizing resource utilization. VDS attempts to service m out of k job instances by their virtual deadlines, that may be some finite time after the corresponding real-time deadlines. Notwithstanding, VDS is capable of outperforming DWCS and similar algorithms, when servicing jobs with potentially different request periods. Additionally, VDS is able to limit the extent to which a fraction of all job instances are serviced late. Results from simulations show that VDS can provide better window-constrained service guarantees than other related algorithms, while still having as good or better delay bounds for all scheduled jobs. Finally, an implementation of VDS in the Linux kernel compares favorably against DWCS for a range of scheduling loads.

Friendly Virtual Machine: Leveraging a Feedback-Control Model for Application Adaptation

With the increased use of "Virtual Machines" (VMs) as vehicles that isolate applications running on the same host, it is necessary to devise techniques that enable multiple VMs to share underlying resources both fairly and efficiently. To that end, one common approach is to deploy complex resource management techniques in the hosting infrastructure. Alternately, in this paper, we advocate the use of self-adaptation in the VMs themselves based on feedback about resource usage and availability. Consequently, we define a "Friendly" VM (FVM) to be a virtual machine that adjusts its demand for system resources, so that they are both efficiently and fairly allocated to competing FVMs. Such properties are ensured using one of many provably convergent control rules, such as AIMD. By adopting this distributed application-based approach to resource management, it is not necessary to make assumptions about the underlying resources nor about the requirements of FVMs competing for these resources. To demonstrate the elegance and simplicity of our approach, we present a prototype implementation of our FVM framework in User-Mode Linux (UML)-an implementation that consists of less than 500 lines of code changes to UML. We present an analytic, control-theoretic model of FVM adaptation, which establishes convergence and fairness properties. These properties are also backed up with experimental results using our prototype FVM implementation.

Measuring the Behavior of a World-Wide Web Server

Server performance has become a crucial issue for improving the overall performance of the World-Wide Web. This paper describes Webmonitor, a tool for evaluating and understanding server performance, and presents new results for a realistic workload. Webmonitor measures activity and resource consumption, both within the kernel and in HTTP processes running in user space. Webmonitor is implemented using an efficient combination of sampling and event-driven techniques that exhibit low overhead. Our initial implementation is for the Apache World-Wide Web server running on the Linux operating system. We demonstrate the utility of Webmonitor by measuring and understanding the performance of a Pentium-based PC acting as a dedicated WWW server. Our workload uses a file size distribution with a heavy tail. This captures the fact that Web servers must concurrently handle some requests for large audio and video files, and a large number of requests for small documents, containing text or images. Our results show that in a Web server saturated by client requests, over 90% of the time spent handling HTTP requests is spent in the kernel. Furthermore, keeping TCP connections open, as required by TCP, causes a factor of 2-9 increase in the elapsed time required to service an HTTP request. Data gathered from Webmonitor provide insight into the causes of this performance penalty. Specifically, we observe a significant increase in resource consumption along three dimensions: the number of HTTP processes running at the same time, CPU utilization, and memory utilization. These results emphasize the important role of operating system and network protocol implementation in determining Web server performance.

On the Interaction Between an Operating System and Web Server

This paper examines how and why web server performance changes as the workload at the server varies. We measure the performance of a PC acting as a standalone web server, running Apache on top of Linux. We use two important tools to understand what aspects of software architecture and implementation determine performance at the server. The first is a tool that we developed, called WebMonitor, which measures activity and resource consumption, both in the operating system and in the web server. The second is the kernel profiling facility distributed as part of Linux. We vary the workload at the server along two important dimensions: the number of clients concurrently accessing the server, and the size of the documents stored on the server. Our results quantify and show how more clients and larger files stress the web server and operating system in different and surprising ways. Our results also show the importance of fixed costs (i.e., opening and closing TCP connections, and updating the server log) in determining web server performance.

Task Assignment in a Distributed System: Improving Performance by Unbalancing Load

We consider the problem of task assignment in a distributed system (such as a distributed Web server) in which task sizes are drawn from a heavy-tailed distribution. Many task assignment algorithms are based on the heuristic that balancing the load at the server hosts will result in optimal performance. We show this conventional wisdom is less true when the task size distribution is heavy-tailed (as is the case for Web file sizes). We introduce a new task assignment policy, called Size Interval Task Assignment with Variable Load (SITA-V). SITA-V purposely operates the server hosts at different loads, and directs smaller tasks to the lighter-loaded hosts. The result is that SITA-V provably decreases the mean task slowdown by significant factors (up to 1000 or more) where the more heavy-tailed the workload, the greater the improvement factor. We evaluate the tradeoff between improvement in slowdown and increase in waiting time in a system using SITA-V, and show conditions under which SITA-V represents a particularly appealing policy. We conclude with a discussion of the use of SITA-V in a distributed Web server, and show that it is attractive because it has a simple implementation which requires no communication from the server hosts back to the task router.

Small-World Internet Topologies: Possible Causes and Implications on Scalability of End-System Multicast

Recent work has shown the prevalence of small-world phenomena [28] in many networks. Small-world graphs exhibit a high degree of clustering, yet have typically short path lengths between arbitrary vertices. Internet AS-level graphs have been shown to exhibit small-world behaviors [9]. In this paper, we show that both Internet AS-level and router-level graphs exhibit small-world behavior. We attribute such behavior to two possible causes–namely the high variability of vertex degree distributions (which were found to follow approximately a power law [15]) and the preference of vertices to have local connections. We show that both factors contribute with different relative degrees to the small-world behavior of AS-level and router-level topologies. Our findings underscore the inefficacy of the Barabasi-Albert model [6] in explaining the growth process of the Internet, and provide a basis for more promising approaches to the development of Internet topology generators. We present such a generator and show the resemblance of the synthetic graphs it generates to real Internet AS-level and router-level graphs. Using these graphs, we have examined how small-world behaviors affect the scalability of end-system multicast. Our findings indicate that lower variability of vertex degree and stronger preference for local connectivity in small-world graphs results in slower network neighborhood expansion, and in longer average path length between two arbitrary vertices, which in turn results in better scaling of end system multicast.

PeriScope: An Active Internet Probing and Measurement API

Growing interest in inference and prediction of network characteristics is justified by its importance for a variety of network-aware applications. One widely adopted strategy to characterize network conditions relies on active, end-to-end probing of the network. Active end-to-end probing techniques differ in (1) the structural composition of the probes they use (e.g., number and size of packets, the destination of various packets, the protocols used, etc.), (2) the entity making the measurements (e.g. sender vs. receiver), and (3) the techniques used to combine measurements in order to infer specific metrics of interest. In this paper, we present Periscope: a Linux API that enables the definition of new probing structures and inference techniques from user space through a flexible interface. PeriScope requires no support from clients beyond the ability to respond to ICMP ECHO REQUESTs and is designed to minimize user/kernel crossings and to ensure various constraints (e.g., back-to-back packet transmissions, fine-grained timing measurements) We show how to use Periscope for two different probing purposes, namely the measurement of shared packet losses between pairs of endpoints and for the measurement of subpath bandwidth. Results from Internet experiments for both of these goals are also presented.

On the Geographic Location of Internet Resources

One relatively unexplored question about the Internet's physical structure concerns the geographical location of its components: routers, links and autonomous systems (ASes). We study this question using two large inventories of Internet routers and links, collected by different methods and about two years apart. We first map each router to its geographical location using two different state-of-the-art tools. We then study the relationship between router location and population density; between geographic distance and link density; and between the size and geographic extent of ASes. Our findings are consistent across the two datasets and both mapping methods. First, as expected, router density per person varies widely over different economic regions; however, in economically homogeneous regions, router density shows a strong superlinear relationship to population density. Second, the probability that two routers are directly connected is strongly dependent on distance; our data is consistent with a model in which a majority (up to 75-95%) of link formation is based on geographical distance (as in the Waxman topology generation method). Finally, we find that ASes show high variability in geographic size, which is correlated with other measures of AS size (degree and number of interfaces). Among small to medium ASes, ASes show wide variability in their geographic dispersal; however, all ASes exceeding a certain threshold in size are maximally dispersed geographically. These findings have many implications for the next generation of topology generators, which we envisage as producing router-level graphs annotated with attributes such as link latencies, AS identifiers and geographical locations.

Sampling Biases in IP Topology Measurements

Considerable attention has been focused on the properties of graphs derived from Internet measurements. Router-level topologies collected via traceroute studies have led some authors to conclude that the router graph of the Internet is a scale-free graph, or more generally a power-law random graph. In such a graph, the degree distribution of nodes follows a distribution with a power-law tail. In this paper we argue that the evidence to date for this conclusion is at best insufficient. We show that graphs appearing to have power-law degree distributions can arise surprisingly easily, when sampling graphs whose true degree distribution is not at all like a power-law. For example, given a classical Erdös-Rényi sparse, random graph, the subgraph formed by a collection of shortest paths from a small set of random sources to a larger set of random destinations can easily appear to show a degree distribution remarkably like a power-law. We explore the reasons for how this effect arises, and show that in such a setting, edges are sampled in a highly biased manner. This insight allows us to distinguish measurements taken from the Erdös-Rényi graphs from those taken from power-law random graphs. When we apply this distinction to a number of well-known datasets, we find that the evidence for sampling bias in these datasets is strong.

A Self-initializing Eyebrow Tracker for Binary Switch Emulation

We designed the Eyebrow-Clicker, a camera-based human computer interface system that implements a new form of binary switch. When the user raises his or her eyebrows, the binary switch is activated and a selection command is issued. The Eyebrow-Clicker thus replaces the "click" functionality of a mouse. The system initializes itself by detecting the user's eyes and eyebrows, tracks these features at frame rate, and recovers in the event of errors. The initialization uses the natural blinking of the human eye to select suitable templates for tracking. Once execution has begun, a user therefore never has to restart the program or even touch the computer. In our experiments with human-computer interaction software, the system successfully determined 93% of the time when a user raised his eyebrows.

Cache-and-Relay Streaming Media Delivery for Asynchronous Clients

We consider the problem of delivering popular streaming media to a large number of asynchronous clients. We propose and evaluate a cache-and-relay end-system multicast approach, whereby a client joining a multicast session caches the stream, and if needed, relays that stream to neighboring clients which may join the multicast session at some later time. This cache-and-relay approach is fully distributed, scalable, and efficient in terms of network link cost. In this paper we analytically derive bounds on the network link cost of our cache-and-relay approach, and we evaluate its performance under assumptions of limited client bandwidth and limited client cache capacity. When client bandwidth is limited, we show that although finding an optimal solution is NP-hard, a simple greedy algorithm performs surprisingly well in that it incurs network link costs that are very close to a theoretical lower bound. When client cache capacity is limited, we show that our cache-and-relay approach can still significantly reduce network link cost. We have evaluated our cache-and-relay approach using simulations over large, synthetic random networks, power-law degree networks, and small-world networks, as well as over large real router-level Internet maps.